Skip to main content

Full text of "Introduction to plant geography and some related sciences"

See other formats

Digitized by the Internet Archive 
in 2013 

Introduction to 

Other Books by the same Author : 

Russian Waters, London: Arnold (1931) 

The Isle of Auks, London : Arnold (1932) 

Botany of the Canadian Eastern Arctic : 

Part I, Pteridophyta and Spermatophyta, Ottawa : King's 
Printer (1940) 

Part II, Thallophyta and Bryophyta (ed.), Ottawa : King's 
Printer (1947) 

Part III, Vegetation and Ecology, Ottawa : King's Printer (1948) 

Arctic Unfolding, London, etc. : Hutchinson (1949) 

Circumpolar Arctic Flora, Clarendon Press, Oxford (1959) 

Arctic Botany (3 vols.), Clarendon Press, Oxford (in Press) 

Editor of Plant Science Monographs, London : Leonard Hill; New 
York: Interscience. 

General Editor of World Crops Books, London : Leonard Hill; New 
York: Interscience. 



and Some Related Sciences 



M.S. (Yale), M.A., D.Phil., D.Sc (Oxon.) 

I'isiting Professor, University of Geneva; lately 
Professor of Plant Ecology and Taxonomy, Head 
of the Department of Botany, and Director of 
the University Herbarium, etc., Faculty of Science, 
Baghdad, Iraq. Formerly Fielding Curator and 
Keeper of the University Herbaria, University 
Demonstrator and Lecturer in Botany, and Senior 
Research Fellozv of Nezv College, Oxford; 
Macdonald Professor of Botany, McGill Univer- 
sity; Research Associate, etc., of Harvard and 
Yale Universities. 


New York Toronto London 



© Nicholas Polunin i960 
First published i960 

Printed in Great Britain by Butler & Tanner Ltd., Frome and London 

Dedicated To 




who inspired and first 
commissioned this book 






Plan of the book 

Geographical patterns 

Climate the master 

The ideal plant . 

Plant sociology . 

The animal side . 

Some earlier works on plant geography 


Classification and nomenclature 



Bryophyta . 


Spermatophyta . 

Further consideration 


Physiological make-up .... 
Ecological limitation .... 
Structural ' adaptations ' of vegetative parts 
Classification by life-forms , 
Further consideration .... 


Wind dispersal . 

Dispersal by water and ice 

Dispersal by animals (apart from Man) 

Dispersal by human agency 

Mechanical dispersal . 


Further consideration . 


Groups of fossil lower plants 
Fossil Seed-plants 

















Past ages and their plant life .... 

Further consideration ...... 



Some effects of relatively recent climatic changes . 
Pleistocene persistence versus subsequent immigration 
Continental drift, shifting poles, land-bridges, etc. . 
Postglacial changes ...... 

The genetical heritage ...... 

Polyploids and their areas ..... 

Further consideration ...... 



* Continuous ' intercontinental ranges 
Discontinuous ranges .... 

Relic areas ...... 

Vicarious areas ..... 

Endemic areas ..... 

Polytopy and the incidence of areas 
Intraneous, extraneous, and other elements 
Major regions ..... 

Further consideration .... 


Effects of cultivation ..... 

Naturalization and acclimatization. 

Some herbaceous crops and their areas 

Forestry and other woody ' crops ' 

Significance and distribution of weeds and plant diseases 

Further consideration ..... 


Foods ..... 

Beverages and flavours 
Medicinals and drugs . 
Fatty oils and waxes . 
Smoking and chewing materials . 
Structural and sheltering materials 
Industrial uses and extractives 
Clothing materials and other fibres 
Fuels (including fossil, etc.) . 
Latexes and exudates . 
Tanning and dyeing materials 
Essential oils and scents (perfumes) 
Insecticides and herbicides . 
Environmental and ecological 
Aesthetic and ornamental 
Microorganisms and miscellaneous 


















Some nuisances . 
Further consideration . 






Biotic . 

Further consideration 


Terrestrial habitats 
Aquatic habitats . 
Alain successions 
Main climaxes 
Further consideration 


Deciduous summer forests . 
Northern coniferous forests . 
Warm-temperate rain forests 
Sclerophyllous, etc., woodlands 
Heathlands and grasslands 
* Semi-deserts and deserts 
Serai communities 
Some physiographic effects 
Further consideration . 



Arctic tundras 

Arctic scrub and heathlands 

Arctic fell-fields and barrens 

Seaside and other local types 

Serai types. 

High altitudes 

Antarctic types . 

Further consideration . 



Tropical rain forests ...... 

Tropical forests with a seasonal rhythm 

Tropical and subtropical savannas and other grasslands 

Semi-desert scrubs ...... 

Tropical and subtropical deserts .... 

Mangrove and other sea-shore vegetation 
Further serai or edaphic communities . 



















Altitudinal effects . . . 

Further consideration ..... 





Some features of the freshwater aquatic environment . 472 

Plankton ........ 

• 479 

Cryophytic communities ..... 

. 489 

Benthos ........ 

. 491 

Bogs and saline waters ..... 


Hydroseres ........ 


Further consideration ...... 

• 506 


• 507 

Some features of the marine environment 


Plankton ........ 

• 515 

Benthic environments and life-forms 

• 521 

Benthos ........ 

• 529 

Aphotic bottoms ....... 

• 538 

Further consideration ...... 

• 540 


• 541 

Landscapes and component landforms . 

• 54i 

Landforms and plant life ..... 

• 543 

Interpretation and uses ..... 

• 55i 

Land-use classification. ..... 

• 555 

Further consideration ...... 

. 561 


ONS 562 

' Adaptations ' of individuals 

• 562 

Man-made adjustments 

• 563 

Vegetational adaptation 

. 565 

Manipulation of vegetation . 

. 566 

Plant geographical study 

• 57i 

Further applicational possibilities 

• 573 

Additional reading 

• 579 

INDEX (including technical terms, which are usually defined 

where first used in the text) . . . . 581 


Generalized land vegetation map of the world 

Facing page i 



















Some plant-like animals ........ 2 

Zoogeographical realms and floristic regions of the world . .18-19 

Various types of Bacteria ........ 26 

Various Blue-green Algae (Cyanophyceae) ..... 28 

Some forms of Green Algae (Chlorophyceae) .... 30-1 

Diatoms (Bacillariophyceae) . . . . . . -33 

Dinoflagellates (Dinophyceae) . . . . . -34 

Some Brown Algae (Phaeophyceae) ...... 36-7 

Various forms of Red Algae (Rhodophyceae) .... 40 

Slime-moulds (Myxomycetes) ....... 42 

Some Fungi .......... 44-5 

Various forms of Lichens (Lichenes) ...... 48 

Types of Liverworts (Hepaticae). ...... 50 

Some Mosses (Musci) ........ 52-3 

Field Horsetail (Equisetiim arvense agg.) . . . . -55 

Types of living Lycopodineae . . . . . . .58 

Various types of Ferns (Filicineae) ...... 60-1 

Some examples of Gymnosperms (Gymnospermae) . . . 64-7 

Features of Angiosperms ........ 70-1 

Features aiding water conservation or absorption .... 83-5 

Features promoting aeration ....... 86-7 

Various adaptations for climbing, twining, scrambling, and running 88-9 
Modifications for storing food or catching insects . . . 90-1 

Diagrams illustrating some Raunkiaer life-forms .... 94 

Wind-dispersal mechanisms and disseminules .... 104-5 

Water-dispersed fruits and other bodies . . . . .110 

Adaptations for dispersal by animals . . . . . .113 

Dispersal by extension of growth or by mechanical propulsion, etc. 122-3 
Some primitive plant fossils ....... 130-1 

Some Psilophytineae . . . . . . . . .134 

Fossil Equisetineae ......... 136 

Fossil Lycopodineae . . . . . . . . 137 

Pteridosperms (reconstructed) . . . . . . .139 

A reconstructed Cycadeoid and parts of living and fossil Ginkgoales 140-1 
Cordaitales (reconstructed) . . . . . . . .142 

Fossil parts of Conifer and Angiosperm . . . . .143 

Distribution of plant groups in geological time as far as known . 145 
Restoration of a late Devonian forest of New York . . .147 

Generalized reconstruction of a Carboniferous forest . . .148 

Reconstructed Triassic landscape . . . . . .150 

A reconstructed scene in Switzerland during Miocene times. . 152 

Known distribution of species of Liriodendron in Tertiary times and 

nowadays . . . . . . . . . .157 

Past and present distributions of Redwoods. . . . .158 

Map showing periods since when various areas of present-day North 

American land have supposedly been free from major ice-sheets, etc. 159 
Maps illustrating ' Continental drift ' and the proximity of land-masses 

as apparently affecting floristic richness .... .168-9 

Map showing arctic circumpolar distribution as exemplified by 

Edwards's Eutrema. ........ 

Maps showing circumboreal and circumaustral distributions . 




Fro. PAGE 

48. Map showing pantropic distribution of the Palm family . .187 

49. Map showing arctic-alpine distribution as exemplified by Saxifraga 

oppositifolia agg. . . . . . . . . .189 

50. Map showing range of Drooping Ladies'-tresses (North Atlantic, 

etc., distribution) ......... 190 

51. Map showing range of Skunk-cabbage (North Pacific distribution) 190 

52. Map showing North-South American distribution of Pitcher-plant 

family . . . . . . . . . . .190 

53. Map showing range of Cimicifuga foetida (Europe-Asian distribution) 191 

54. Map showing range of species of Platanus . . . . .192 

55. Map showing pantropical discontinuous range of the genus Buddleia 

and the mainly neotropical range of the family Vochysiaceae . 193 

56. Map showing range of the genus Jovellana (South Pacific distribution) 194 

57. Map showing range of the genus Asclepias (South Atlantic, etc., 

distribution) . . . . . . . . . .194 

58. Map showing (Antarctic) range of the genus Nothofagus . .195 

59. Map showing (bipolar) range of the genus Empetrum . . .196 

60. Map showing intracontinental discontinuous distribution in Australia 

of the section Erythrorhiza of the genus Drosera . . .197 

61. Map showing ' Lusitanian ' distribution of Mackay's Heath . .197 

62. Maps showing known localities of Low Sandwort . . .198 

63. Map showing recent and 'fossil' stations of the Water-chestnut in 

Scandinavia . . . . . . . . . .199 

64. Transatlantic vicariads ........ 203 

65. Map showing main vegetational-climatic regions of the world . 213 

66. Geographical distribution of world Rice production . . . 225 

67. World Wheat production ........ 227 

68. World Rye production ........ 228 

69. World Maize production ........ 230 

70. World Potato production ; also some Greenland vegetables . .232-3 

71. World Flax-seed production ....... 236 

72. World Cotton production ........ 237 

73. World Peanut (Groundnut) production . . . . .238 

74. World Tobacco production. ....... 240 

75. World distribution of annual Cane and Beet Sugar production . 241 

76. World production of Cocoa Beans ...... 244 

77. World Coffee production ........ 245 

78. Distribution of the world's human population .... 256 

79. Moss Campion flowering only on the south-facing sides and tops of 

its domed tussocks near 8o° N. in Spitsbergen .... 285 

80. Annual mean, and mean temperature of the warmest month, in 

different parts of the world ....... 286-7 

81. Average annual precipitation in different parts of the world . . 290 

82. Effect of wind on trees ........ 292 

83. Aspect effects in Colorado and Nepal ...... 295 

84. Three soil profiles of comparable depth ..... 299 

85. Distribution of primary soil groups of the world .... 300 

86. Some important effects of grazing ...... 308 

87. Devastating results of overgrazing ...... 309 

88. Margin of tropical oligotrophic lake, with steep rocky sides and rapidly 

deepening water, supporting few larger plants . . . 318 

89. Lake of eutrophic type near Prout's Neck, Maine . . .319 

90. Diagram illustrating stages of hydrosere with deposition of peat . 326 

91. Sedge-meadow stage of hydrosere colonized by some hygrophytic 

shrubs and trees, near Prout's Neck, Maine . . . .326 

92. Telescoped stages of xerosere in Norwegian Lapland . . .327 

93. Psammosere at Prout's Neck, Maine, showing Marram Grass binding 

sand above high-tide mark ....... 328 

94. Leafless condition of mixed deciduous summer forest in northeastern 

United States — in winter . . . . . . -338 


95. Mixed deciduous forest in northeastern United States — in summer 339 

96. Open scrubby Birch ' forest ' in Finmark, northern Norway . . 344 

97. Boreal coniferous forest surrounding lake-side bog in sheltered valley 
in Troms, northern Norway . . _ . . . . . 347 

• 347 

• 349 

• 35o 

• 352 

• 353 

• 356 

• 356 

98. Outside the taiga in northern Ungava, Canada 

99. ' Lake-forest ' of southeastern Canada, in summer 

100. The same area of ' lake-forest ' but under winter conditions 

snow covering the ground ..... 

1 01. Warm-temperate rain forest in southeastern United States 

102. Bald-cypress swamp in southeastern United States 

103. Rocky area with patches of mixed scrub of ' maquis ' type 

Mediterranean island of Corsica .... 

104. Sierran chaparral climax, Santa Barbara, California 

105. Sclerophyllous forest in Australia, dominated by lofty Gum-trees . 357 

106. The North American Prairie : mid- and short-grass communities. 362 

107. Salt desert and salt-marsh of a warm-temperate region . . 367 

108. Pine ' Krummholz ' at timber-line in the Rocky Mountains . -377 

109. Tundra on Southampton Island, Hudson Bay .... 384 
no. Marshy tundra near the south shore of Hudson Strait, dominated by 

Cotton-grasses, Sedges, and Grasses ..... 385 

in. Dry tundra on raised area overlooking Hudson Bay . . .386 

112. Mesophytic Sedge-grass tundra on the north coast of Alaska near 

Point Barrow, overlooking the Arctic Ocean . . . .388 

113. Extensive area of damp ' hillock tundra ' on the coast of Spitsbergen 388 

114. Discontinuous tundra-like tract of mixed Grasses, Northern Wood-rush, 

forbs, and Polar Willow, in inland valley, West Spitsbergen, grazed 

by a pair of wild Reindeer . . . . . . .389 

115. Tangled Willow scrub in the low-arctic belt of the Northwest 

Territories, Canada. ........ 391 

116. Patchy scrub of Glaucous Willow and Scrub Birch, up to nearly 

2 metres high, in southwestern Greenland . . . .391 

117. Dense low-arctic heath dominated by Arctic Blueberry in northern- 

most Quebec .......... 393 

118. 'Snow-patch' darkened by Arctic Bell-heather, southern Baffin 

Island ........... 394 

119. Mixed middle-arctic heath with many light-coloured and other 

Lichens ........... 394 

120. ' Polygons ' in northernmost Spitsbergen ..... 396 

121. Fell-field on calcareous soil in exposed situation, northernmost 

Labrador .......... 397 

122. Lichen barrens in the uplands of central Baffin Island, looking south 398 

123. Monotonous tract of prairie-like fell-field in the vicinity of the 

Magnetic North Pole, Prince of Wales Island, Canadian Arctic 
Archipelago . . . . . . . . . .398 

124. Shingly beach bound by swarded Lyme-grass .... 400 

125. Looking down on a salt-marsh dominated by Pacific Silverweed and 

Creeping Alkali-grass ........ 400 

126. Luxuriant ' patchwork quilt ' of mixed and many-coloured Lichens 

and Mosses developed near top of bird-cliff . . . .401 

127. Looking down on a luxuriant mossy mat on the manured periphery 

of a wildfowl nesting-ground in Spitsbergen .... 402 

128. Purple Saxifrage barrens on exposed ridge overlooking the sea in 

northernmost Baffin Island ....... 403 

129. ' Late-snow ' patch in the highlands of central Baffin Island . 404 

130. Looking down on the Herb-like Willow zone of the late-snow area 

shown in Fig. 129 . . . . . . . . . 404 

131. Luxuriant lakeside marsh dominated by Water Sedge and Tall 

Cotton-grass .......... 406 

132. Fine bed of Scheuchzer's Cotton-grass beside tarn in northern 

Spitsbergen .......... 407 



133. Top of flower-slope below weathering crag in southern Baffin Island 408 

134. Spitsbergen flower-slope dominated by Alpine Arnica . . . 408 

135. High in the mountains near the margin of the ice-sheet in southern 

Greenland . . . . . . . . 411 

136. Upland scrub of Dwarf Birch and silky-leafed Willows constituting 

an altitudinal climax above tree-limit in northern Norway 411 

137. High-alpine vegetation and flowering . . . . .412 

138. Alpine puna-like formation near mountain summit in Colombia 414 

139. Crustaceous and foliose Lichens on rocks near shore, Goudier Islet, 

Antarctica . . . . . . . . . .416 

140. Luxuriant growth of Mosses, broken chiefly by rocks bearing Lichens, 

extending up snow-melt gully in area frequented by Penguins, 
Deception Island, Antarctica . . . . . . .418 

141. Profile diagram of primary mixed tropical rain forest, Moraballi 

Creek, British Guiana ........ 426 

142. Profile diagram of climax evergreen forest in Trinidad, British West 

Indies ........... 426 

143. Tropical rain forest in the Philippine Islands .... 427 

144. Another scene of tropical rain forest in the Philippines . 429 

145. Base of tree-trunk showing exaggeratedly buttressed roots in tropical 

rain forest .......... 429 

146. An epiphytic Fern which has small humus-gathering leaves and 

larger photosynthetic ones that also produce spores . . -433 

147. An epiphytic Bromeliad with a mass of fibrous roots investing the 

branch of the ' host ' tree . . . . . . -433 

148. Roots of Strangling Fig on a large tree-trunk . . . -435 

149. An old specimen of Strangling Fig in which the roots serve as trunks, 

the original ' host ' having disappeared . . . . .436 

150. Flower and buds of Rafflesia manillana, a true parasite on the roots of 

a Cissus vine .......... 437 

151. A tropical hemiparasitic Mistletoe, Viscum orientale, the root of 

which forms a single haustorium ...... 438 

152. Palm-savanna in southern Florida ...... 445 

153. Savanna in Australia under rainfall of 25-75 cm. annually . 446 

154. Semi-desert ' bush-land ' in Australia ...... 448 

155. Arizona near-desert scene showing the giant Saguaro (or Sahuaro) 

Cactus and bushy Ocotillo . . . . . . .451 

156. Areas of desert in central Iraq ....... 453 

157. A typical Mangrove plant, Rhizophora candelaria, forming a char- 

acteristic marginal ' mangrove ' and showing prominent prop-roots 
below . . . . . . . . . . .455 

158. Interior of Philippine mangrove-swamp forest at low tide . . 455 

159. A Screw-pine and a subtropical estuarine salt-marsh . . . 459 

160. Coconut Palms along a tropical sea-coast . . . . .461 

161. Two-storied montane rain forest at an altitude of 740 metres in the 

Philippine Islands ......... 468 

162. Epiphytes on trunk of tree near upper limit of montane rain forest in 

the Philippine Islands ........ 4^9 

163. Mossy elfin forest near summit of mountain, Ihilippine Islands 470 

164. Vascular plants floating freely on fresh water . . . .484-5 

165. Diagrammatic representation of the plankton in a Wisconsin lake 

during May to October ........ 486 

166. Diagram indicating distribution in a European lake of a cyanophyeean 

(Glosotrichia echinulata), which is rendered buoyant by included 
gas-vacuoles .......... 487 

167. Diagrammatic representation of a typical lake-marginal profile . 493 

168. Diagram of cross-section through a ' highmoor ' bog that has arisen 

from a small lake . . . . . . 501 

169. Leaves of Sacred Lotus projecting out of the water, and Pistia stratiotes 

floating on the water, in the Philippine Islands . . 504 




170. Diagrammatic representation of typical sea-marginal profile . 

171. Photomicrographs of marine phytoplanktonic communities 

172. Postelsia palmaeformis . ....... 

173. Scene at low tide on a rocky sea-shore of the eastern United States 

174. A characteristic Kelp, Alaria dolichorhachis .... 

175. A giant Pacific Kelp, Macrocystis pyrifera .... 

176. An arctic foreshore photographed from near low-tide mark . 

177. Old sand-dune colonized by shrubs and Pitch Pine after stabilization 

by Marram Grass ........ 

178. An example of Class VII land ...... 

179. An example of Class VIII land ...... 

180. Illustration of the eight land-use capability classes 

181. Part of a maze of gullies which crosses more than an entire county 

in the southern United States ....... 

182. Destructive water-erosional gully in heavily overgrazed pasture in 

Illinois .......... 

183. The same as Fig. 182, two years later .... 

184. Map showing positions of meteorological stations of the Ukraine, 

with indications of their climatic analogues in the United States 









To one who considers that personal friendship and generosity 
in sharing the fruits of scholarship are among the very best things 
of life, it gives great pleasure to acknowledge indebtedness to the 
many savants who have contributed of their knowledge or store of 
illustrations to the benefit of this work. They must not, however, 
be held responsible for any shortcomings it may have — such as, 
perhaps, in some views, omission of discussion of certain con- 
troversial issues which seemed best by-passed at least at the time 
of writing. Notable among these are Dr. C. W. Thornthwaite's 
work on evapotranspiration and the classification of climates, 
Professor Eric Hulten's views on the history of arctic and boreal 
species, and various ideas about the places of origin of plant forms 
and the possibilities of their being multiple (polytopic). 

The book owes its inception to the foresight of Dr. George H. T. 
Kimble, who, when Director of the American Geographical Society, 
was instrumental in my being invited and generously commissioned 
to prepare it for a new series of ' readers ' on geographical subjects. 

Early on, the general plan was improved from time to time 
following discussion with colleagues at Oxford, Yale, and Harvard 
Universities, in its near-final form being approved by a seminar at the 
last-named. The plan also derived benefit from many individuals 
elsewhere ; among these my former teacher and chief, the late 
Professor Sir Arthur G. Tansley, and my former pupils, Professor 
John H. Burnett and Dr. John Warren Wilson, were particularly 
helpful. Yet others who made valuable suggestions, most of which 
were gladly adopted, include Professor Hugh M. Raup of Harvard 
University, Professor Paul B. Sears of Yale University, Professor 
Joseph Ewan of Tulane University, and Drs. Raymond F. Fosberg 
and Henry K. Svenson, both of Washington, D.C. The then 
Directors of the two main botanical gardens of the United Kingdom, 
the late Professor Sir William Wright Smith of Edinburgh and Sir 
Edward J. Salisbury of Kew, also gave freely of their advice, as did 
the former Director of the New York Botanical Garden and of the 
Arnold Arboretum, the late Professor Elmer D. Merrill. The book, 
moreover, derives much from the able (and direct) teaching of two 



others who are unfortunately no longer with us, namely the late 
Professors George E. Nichols of Yale (in ecology) and Merritt L. 
Fernald of Harvard (in taxonomy). 

Colleagues or friends who have been kind enough to read and 
give helpful advice about particular chapters or groups of chapters 
have included Professors G. E. Hutchinson, Harold J. Lutz, Paul 
B. Sears, and Mr. Albert F. Burke, all of Yale University, Pro- 
fessors Kenneth V. Thimann and Hugh M. Raup, and Drs. A. F. 
Hill and Richard E. Schultes, all of Harvard University, Drs. W. O. 
James, F.R.S., and F. H. Whitehead, both of Oxford University, 
Drs. H. Hamshaw Thomas, F.R.S., and Harry Godwin, F.R.S., 
both of Cambridge University, the late Professor Sir Arthur G. 
Tansley, F.R.S., of Grantchester, Cambridge, Professor Paul W. 
Richards of the University College of North Wales, Bangor, Dr. 
John Hutchinson, F.R.S., of the Royal Botanic Gardens, Kew, Mr. 
F. T. Walker of the Institute of Seaweed Research, Inveresk, 
Musselburgh, Scotland, Messrs. Robert Ross and W. T. Stearn, both 
of the British Museum (Natural History), Dr. Richard S. Cowan of 
the New York Botanical Garden, Professor G. W. Prescott of 
Michigan State University, Professor George L. Church of Brown 
University, Professor Valentine J. Chapman of Auckland University 
College, New Zealand, Professor G. Einar Du Rietz, of Uppsala, 
Professors Gunnar Erdtman of Stockholm and Karl H. Rechinger 
of Vienna (while Visiting Professors in my department at Baghdad, 
Iraq), Professor Thorvald Sorensen, of Copenhagen, Professor John 
H. Burnett, now of the University of St. Andrews, Dr. Frank E. 
Egler of Aton Forest, Norfolk, Conn., and Dr. Edward H. Graham, 
Director of Plant Technology in the Soil Conservation Service of 
the United States Department of Agriculture, Washington, D.C. 
Whereas the choice of these kind mentors was naturally governed 
largely by their specialist interests, to mention who ' passed ' what 
might leave them open to being held responsible for errors of com- 
mission or omission which are in fact my own. Most of the sub- 
stance of this book was earlier presented in a full-year graduate 
course at Yale University — a circumstance which, at the instance of 
some senior participating students, has led to further constructive 
comment and, surely, improvement. 

In the matter of illustration, so vitally important to a work of 
this kind, the greatest debt is to Ginn and Company, of Boston, 
Massachusetts, and Mrs. William H. Brown, for their loan of, and 
permission to use freely, so many of the fine drawings and photo- 


graphic prints made for the late Professor William H. Brown's The 
Plant Kingdom, published in 1935. This was not only a great 
convenience but also a great privilege, these illustrations being often 
of unsurpassed excellence. Acknowledgment is also due to the 
National Museum of Canada for permission to reproduce many of 
my photographs of arctic regions that are now in their possession. 
Other sources of illustrations, where not contributed by myself, are 
acknowledged individually. 

Nicholas Polunin 
Faculty of Science, 

Baghdad, Iraq 
Spring, 1957 * 

1 Since this was written it has not been possible to incorporate extensively any 
new ideas or to consider subsequent works, though some details of publication 
have been brought up to date. The proofs have kindly been read by Professor 
John H. Burnett and Dr. A. D. Q. Agnew (now of my Department in Baghdad), 
while in connection with their correction warm thanks are due to my secretary, 
Miss Christine Wright. Acknowledgment is also made to Dr. B. Barnes for his 
valued part in the preparation of the Index. 

Chapter I 

Let us begin with a few basic definitions and follow them with 
some general explanations. 

Biology is the science of life, the study of living things, and it 
has two main branches — botany, which deals with plants, and 
zoology, which deals with animals. But whereas every one of us 
must surely be clear about the differences between the typical plant 
(which is static, green, and does not ingest solid food) and the 
typical animal (which is motile, not green, and ingests elaborated 
food), there remain many ' border-line cases ' that are apt to be 
claimed by both botanists and zoologists. Indeed, each of the 
characteristics mentioned for one of these primary groups (king- 
doms) of living organisms is exhibited by some members of the 
other, which prevents the drawing of any hard and fast line between 
all animals and all plants. And even if we add the stipulation that 
the greenness of plants shall be due to chlorophyll, and that they 
shall contain the carbohydrate cellulose, there remain many organisms 
which possess neither feature but still in other ways seem to be 
plants, and are usually treated as such. 

Consequently it seems best in this case not to attempt precise 
definition but rather to visualize the typical plant as a living organism 
that is fixed, possessed of cellulose cell-walls, and dependent for its 
main food-supply upon simple, gaseous or liquid substances (princip- 
ally carbon dioxide and water). With the aid of chlorophyll in the 
light, the organism builds up these substances into sugars and other 
complex materials. The green plant is thus responsible for the 
fundamental chain of reactions on which almost all life depends. 
But this partial description excludes many organisms (such as Yeasts 
and other small Fungi) which are commonly considered to be plants. 
These ' exceptions ' often form major groups although, as we shall 
see in the next chapter, they may exhibit none of the stipulated 
main plant characteristics. The description also leaves behind a 
basic ' hub ' of organisms, chiefly of microscopic types, that seem 
to belong almost as much to one kingdom as to the other. Among 


Chapter I 

Let us begin with a few basic definitions and follow them with 
some general explanations. 

Biology is the science of life, the study of living things, and it 
has two main branches — botany, which deals with plants, and 
zoology, which deals with animals. But whereas every one of us 
must surely be clear about the differences between the typical plant 
(which is static, green, and does not ingest solid food) and the 
typical animal (which is motile, not green, and ingests elaborated 
food), there remain many ' border-line cases ' that are apt to be 
claimed by both botanists and zoologists. Indeed, each of the 
characteristics mentioned for one of these primary groups (king- 
doms) of living organisms is exhibited by some members of the 
other, which prevents the drawing of any hard and fast line between 
all animals and all plants. And even if we add the stipulation that 
the greenness of plants shall be due to chlorophyll, and that they 
shall contain the carbohydrate cellulose, there remain many organisms 
which possess neither feature but still in other ways seem to be 
plants, and are usually treated as such. 

Consequently it seems best in this case not to attempt precise 
definition but rather to visualize the typical plant as a living organism 
that is fixed, possessed of cellulose cell-walls, and dependent for its 
main food-supply upon simple, gaseous or liquid substances (princip- 
ally carbon dioxide and water). With the aid of chlorophyll in the 
light, the organism builds up these substances into sugars and other 
complex materials. The green plant is thus responsible for the 
fundamental chain of reactions on which almost all life depends. 
But this partial description excludes many organisms (such as Yeasts 
and other small Fungi) which are commonly considered to be plants. 
These ' exceptions ' often form major groups although, as we shall 
see in the next chapter, they may exhibit none of the stipulated 
main plant characteristics. The description also leaves behind a 
basic ' hub ' of organisms, chiefly of microscopic types, that seem 
to belong almost as much to one kingdom as to the other. Among 


the more important of these are the Bacteria, which cause so many 
of our worst diseases but in other instances benefit us greatly. 
These and other ' border-line cases ', which include many of the 
most primitive organisms living today, will be considered as within 
our immediate sphere of interest. 

Fig. i illustrates some cases of plant-like animals ; several 
animal-like plants will be described and illustrated in the next 


-Some plant-like animals. A, Hydra ( < 20); B, Obelia (X about 12): 
C, a Sponge (x about $); D, a Coral (x %). 

chapter. Defining and classifying such nebulous groups is one of 
the trials and at the same time one of the fascinations of biology. 
Geography is the study and description of the differentiation 
and distribution of earthly phenomena, embracing all that composes 
or affects the earth's surface — including its physical features, climates, 
and products whether living or inert. A major branch is biological 
geography, or biogeography, which for practical purposes is usually 
subdivided along the main line of division of living things into two 
kingdoms, so yielding plant geography and animal geography. Our 
main subject, plant geography, also called phytogeography (from 


the Greek word cpvxov, a plant), accordingly deals with the plant 
cover of the world — with its composition, its local productivity, and 
particularly its distribution. This matter of distribution should be 
tackled both on the separate basis of individual species, etc., and 
collectively by dealing with their various and complex assemblages 
that make up vegetation. Our object will be to describe and inter- 
pret all we can of the manifestations of plant geography, paying 
special attention to the differences and similarities existing between 
the various floras and vegetations of the world. The continued 
increase in total human population makes such a study vitally 
significant, Man being dependent on plants for the very where- 
withal of his existence. 

The e conomic imp orta nce of our subject stems from the fact tha t 
gree n plants alone, on any substantial scale, are able to build, from 7 po> 
sImpieTawjnat erials and e nergy d erived from sunlight, the comple x 
substances on which animals as well as the plants themselves all 
depend for food. This food constitutes the main source of material 
used in body-building, and in it is locked the energy required for 
the various processes of life. Animals, with their usually active 
existence, commonly need this energy in abundance. In them, as 
in plants, it is liberated by the process of respiration, which is a 
kind of slow burning that takes place in living matter and gives to 
Mammals and Birds their bodily heat. Green plants provide food 
for us directly, when we eat them or their products, or indirectly, 
when we eat animals that fed on plants or were at least ultimately 
dependent upon some form of plant life, as indeed all are. 

Plants also provide us with much of our clothing and housing as 
well as industrial raw materials, while in the world as a whole they 
largely condition our environment — forests, for example, being clearly 
different to live in from grassy plains or desert oases. Indeed, 
many of the major migrations of Man and other animals have been 
primarily due to plant distribution. Plants constitute for mankind 
the main inexhaustible source of fuel and industrial supplies and 
are of fundamental importance in many different branches of 
industry : in drug production, brewing, pulp and paper making ; 
in lumbering, in the textile industries, in tanning, dyeing and curing ; 
in the production of plastics, animal feedstuffs, scent, oil, rubber, 
resin, gum, wax and fibres ; and, of course, in the wider fields of 
agriculture, horticulture, forestry, fish-culture, and the direct uses 
of their innumerable products. 

It can be seen from the contents of the earlier works listed at 


the end of this chapter that authorities have differed greatly in the 
past as to the bounds and, in their view, legitimate content of the 
science of plant geography. In the present introductorv work it 
will be interpreted in a much wider sense than usual, as including 
not only all geographical manifestations of plants whether single or 
collective (and hence a good deal of economic and some morpho- 
logical botany), but also the reasons behind these manifestations. 
This will presuppose some consideration of the bases of distributions 
in space and time, and consequently of the relationship to environ- 
ment (ecology), of the classification and systematic arrangement of 
different kinds of plants particularly through their external form 
(taxonomy and systematy), of the study of their internal workings 
(physiology), of their economic importance, and of other disciplines 
that are not normally thought of as plant geographical — hence in 
part the reference to ' some related sciences ' in the sub-title of 
this book, to certain of which in some modest degree it may also 
serve as a general introduction. 

The ultimate purpose of geography is the study of the differences 
in the areas which make up the world. Yet when the plant popula- 
tions are taken into consideration it comes as no surprise, in view 
of their extreme variability, to find that one of the main results of 
such a study is the realization that each area is unique. Once we 
leave the ' systematic ' study of particular phenomena, such as the 
relationship of individual kinds of plants to different areas, and 
enter the ' regional ' sphere of correlation of the various manifesta- 
tions which point to these differences in area, the problem of 
organizing our study becomes almost overwhelming. With any set 
of phenomena as infinitely variable as vegetation (in both time and 
space, as we shall see), the areal integration desired in their geography 
is rendered practicable only by ignoring variations within the smaller 
unit-areas, which may then be studied together and ' lumped ' into 
larger ones. 

Plant geography attempts to integrate these floristic and vegeta- 
tional features as far as possible on a world basis, and for recording 
and illustration makes use of maps as one of its main tools. But the 
very construction of these maps presupposes the utmost care in the 
comparison of the entities whose ranges they indicate. Lack of 
such care is one of the greatest limitations with which the plant 
geographer is faced. Another is the still fragmentary state of Man's 
knowledge of the distribution of the vast majority of the many 
hundreds of thousands of different kinds of plants inhabiting the 


world — not to mention their proper delimitation and description. 
Yet another limitation is the extreme difficulty of collating such 
complex and often distant ' entities ' as vegetation-types, with all 
their infinite variation and intricate intergradation. Even so, when 
no more research than has already been accomplished along these 
lines is brought together, we have a very impressive volume of 
material from which it seems permissible to make some useful 
generalizations, and on which we can build further our edifice of 
plant geography. 

Plan of the Book 

As plants are our chief concern, we shall, after the present intro- 
ductory chapter, first describe the main groups into which the 
myriad forms comprising the plant kingdom (in the wide sense) are 
classified. For each group we shall give some account of how its 
members live and reproduce, with mention of their habitats, dis- 
tributions, and individual importance, and illustrations of examples. 
Then we shall have at least some conception of what we are dealing 
with, and, if previously uninitiated, have an opportunity of becoming 
familiar with our tools. 

Our other main concern being with geographical phenomena and 
particularly with area, we shall deal next with the physiological 
attributes and external features that enable particular plants to grow, 
or prevent them from living, in particular circumstances. In this 
third chapter we shall also touch on the subject of plant classification 
by means of the various ' life-forms ' which are brought about largely 
by the nature of the environment. This is particularly important 
because the reactions of plants to the environments in which they 
exist (see pp. 8-9) constitute one of the main ' keys ' to their 
geographical ranges. In the next chapter we shall consider the means 
by which plants disperse themselves and migrate— with the kinds 
of aids they employ and, incidentally, some of the hindrances they 
meet in attaining their present-day distributions. The following 
chapter, our fifth, will deal with the early evolutionary development 
of plants, and the sixth will be concerned particularly with those 
developments in recent geological ages which have most profoundly 
influenced plant distributions as we see them nowadays. 

In Chapter VII we shall go on to consider examples of the main 
types of distribution and consequent areas recognized today, where 
possible interpreting them in the light of information contained in 


the earlier chapters. This consideration of ' natural ' distributions 
will be followed by a chapter on man-made ones — both intentional 
(of crops) and unintentional (of weeds, etc.). And as the crops, or 
potential crops, of the various parts of the world introduce some 
of the greatest problems of mankind today, the next chapter will 
emphasize the economic, and basic, significance of plant life. For 
the geographical ranges of plants important to Man are often largely 
determined by him, on whom their very existence may depend. 

In these initial chapters we will be dealing chiefly with special 
kinds or systematic groups of plants and their distributions. This 
is little more than a prelude to consideration of the natural group- 
ings, the complex and variable assemblages of different plant com- 
munities each composed of more or less numerous and diverse kinds 
of plants, that make up collectively what we term vegetation. Before 
actually beginning our study of vegetation we must consider the 
environmental conditions (the ecological factors) which largelv 
control its distribution and form : such consideration will occupy 
Chapter X. Some attention will also be given to physiological 
make-up, which primarily determines the reaction of a plant to its 

The ecological factors at a point collectively constitute the habitat, 
or ' habitat conditions '. The habitat, the place where an organism, 
or commonly many organisms, live, may vary greatly from place 
to place but tends to recur in at least comparable form in manv 
different places. Particular habitats are often relatively uniform 
over considerable areas, as in the cases of salt-marshes, shallow 
ponds, and sandy plains. Moreover, when a bare or disturbed area 
is left alone, the vegetation inhabiting it tends to change, exhibiting 
a series of vegetational types ranging from the first lowly colonists 
to a relatively stable community which is ultimately the highest 
the area can support, the advancing series being called a ' succession ' 
or ' sere ', and its outcome the ' climax '. Chapter XI will deal 
in a general way with the main types of plant habitats, successions, 
and climaxes to be distinguished. 

The next five chapters will outline and illustrate the chief vegeta- 
tional types of the world, starting with those to be observed in 
temperate and adjacent lands as being most familiar and compre- 
hensible to the greatest number of us. Following an account of 
the vegetational types of polar lands and high altitudes elsewhere, 
will be a chapter on tropical and adjacent lands, and thereafter one 
on the plant communities of fresh and inland saline waters, wherever 


they may be located, and another (Chapter XVI) on the communities 
of the oceans and seas. 

Landscapes are often notable for peculiarities which concern the 
vegetation more than the actual surface of the land, and in Chapter 
XVII we will consider how landforms, which make up landscapes, 
tend to be characterized in this way. The study of vegetation, 
especially of the more established communities, generally gives a 
better indication of the combined action of environmental factors 
than all manner of measurements. Consequently the plant geo- 
graphical and ecological evidence afforded by an area can be of 
the greatest practical value in interpreting local conditions and in 
planning the best use of land — particularly for agriculture and 

The concluding chapter deals with (i) some natural adaptations 
and (2) man-made adjustments, both in (a) individual plants and 
(b) vegetation. This consideration gives us by cross-inference four 
sets of topics, all of great interest and importance. Examples of 
these are (id) evolution and its mechanisms, (ib) successional 
change in vegetation, (2d) plant-breeding, and (zb) combating 
erosion (itself usually brought on by Man's desecration of vegeta- 
tion). The final paragraphs survey some of the more useful methods 
of study of plant geography, and give further indication of the 
values and future possibilities of the subject — both academically and 
in the service of Man. Despite vast advances in recent decades, 
there remain whole hosts of unsolved problems ; and, indeed, it is 
to be questioned whether this last chapter can ever be brought to 
a satisfactory closure. For such is biological science — an unending 

Most chapters conclude with some indication, in smaller type, of 
how further pertinent information may be obtained through recom- 
mended books which are cited for the purpose. Shorter contribu- 
tions are ignored in this connection as being too numerous and 
difficult to select, as well as usually unavailable to the layman, 
although naturally much of the material presented in this book has 
been drawn from such specialist ' papers \ 

Geographical Patterns 

The botanical aspects of what may be termed geographical or 
areal patterns constitute much of plant geography. Such plant- 
distributional patterns are partly based on physiological reaction to 


ecological factors and, consequently, to a considerable extent on 
climate (see next section). An understanding of them is fundamental 
to our main subject, so some consideration of them seems desirable 
at this stage. 

Just as the land-masses of the world, for example, make up a 
definite (if seemingly unorganized) pattern on the surface of the 
globe, so do other features, that are likewise definable in area, make 
up their own special patterns. Such pattern-forming features 
include the various factors of the environment, w r ith which we shall 
deal in Chapter X. Thus, certain ranges of temperature, for 
example, obtain only within certain areas, and the same is true 
especially of other climatic features (see next section). 

In the simplest case it might be supposed that a particular kind 
of land-plant, needing land to live on, could occupy all of the water - 
and ice-free land of the globe. But in actual fact quite numerous, 
often interdependent and overlapping, environmental and other 
factors prevent this, and no known kind of plant, however wide its 
habitat tolerance, occupies more than a very small proportion of the 
world's surface. At the other extreme are the numerous species 
which appear to inhabit only one limited tract of the globe or even 
a single spot. Each and every species has its particular area, its 
geographical distribution, whether this be small or large, and 
whether continuous or broken up into a more complicated pattern. 
And the pattern will be related to some particular factor or factors 
of the environment, to some migrational ability the plant may 
possess, and/or to evolutionary and geological or more recent 

These migrational tendencies, together with the historical aspects 
of distribution, will be discussed in Chapters IV, V, and VI, and 
it will be found that such aspects may greatly affect the areas at 
present occupied by particular plants. But the factors of the 
environment are apt immediately to limit and circumscribe the area 
that can be occupied by a plant and, although treated in fair detail 
in a later chapter, require some explanation here before we can 

Any condition of the habitat, whether climatic, physiographic, 
edaphic (concerned with the soil), or biotic (concerned with living 
organisms), may limit the area occupied by a plant, and usually 
many of these conditions do come into play. A simple instance is 
that a tropical plant requires warm conditions — or at least, it cannot 
grow in the cold. Usually, however, matters are far more com- 


plicated than this, in that such a plant commonly requires also 
environmental conditions within a certain range of moisture, light, 
soil type, etc. As the world affords, usually over considerable areas, 
practically every conceivable combination of habitat factors, the 
area occupied by a particular plant will, inter alia, depend upon its 
physiological make-up and reaction to the component factors. 
Herein lies the close connection between particular plants and their 
favoured habitats. 

Whereas probably no two situations or even areas of a seemingly 
uniform habitat are exactly identical, we must in practice accept 
them as being alike, even as they appear to be so accepted by plants. 
Accordingly the similar habitats of the world may be grouped together 
to form recognizable patterns ; and, looking at things the other way, 
we find that there usually is a pattern of areas occupied by each 
plant. Often a single habitat factor lies behind such a pattern of 
plant distribution and may readily be recognized as doing so. But, 
unfortunately for those who crave simple monistic explanations, the 
area potentially inhabitable by a particular kind of plant, as pre- 
scribed by suitable habitat conditions, and that area which it actually 
occupies in the world today, are rarely if ever identical or even 
similar. Yet although the area which might be occupied is of both 
interest and importance to the scientist and to mankind, the plant 
geographer's immediate concern is with the fact, i.e. the actual area 

In the same way, the many different types of vegetation form 
geographical patterns of their own, but here again the plant geo- 
grapher is concerned more with the effects, i.e. the actual patterns, 
than with the causes which properly belong to the historical side of 
his studies. The vegetation pattern is to a considerable extent the 
sum of the overlapping distributions of the component plants ; but 
it commonly has a form of its own, for in biology the sum total 
of the components does not necessarily constitute the expected 
whole. The organisms' interrelationships and reactions add much 
that is new to the system and form an integral part of the end result. 

Climate the Master 

As climate tends to supply the most important over-all factors 
determining plant distribution, it behoves us to give at this stage 
some outline of the main ' world ' types of climate and related 
vegetational features. Further details on climatic factors will be 


found in Chapter X, with accompanying figures indicating tempera- 
ture and precipitation in different parts of the globe, and in such 
recognized works as W. G. Kendrew's Climatology, second edition 
(Clarendon Press, Oxford, pp. xv + 400, 1957). The main vegeta- 
tional types occurring on land are dealt with particularly in Chapters 
XII, XIII, and XIV. 

Climate is the most far-reaching of the natural ' elements ' 
controlling plant life, and its study, climatology, is accordingly 
fundamental to plant geography and related disciplines. In the 
words of Kendrew (I.e.), ' "Climate" is a composite idea, a generaliza- 
tion of the manifold weather conditions from day to day throughout 
the year. ... In the study of climatology the primary interest 
lies in the facts of the climates of the earth in themselves, and as 
elements in the natural environment of life.' To the phrase 
* throughout the year ' the ' historical ' plant geographer might wish 
to add ' and through the ages '. 

Climatology deals with the atmospheric conditions which affect 
life — particularly light, temperature, precipitation, evaporating power, 
and wind. Additional factors include radiation, cloudiness, and 
storms. These components are often interdependent, their various 
combinations giving us the characteristic climates of different parts of 
the world which for our purposes may be divided broadly into three. 
These are the polar, temperate, and tropical regions, and they 
are primarily temperature zones. For convenience, the temperate 
areas lying north and south of the equator are considered together, 
as are the north and south polar areas in their turn. In this book 
the temperate regions are purposely treated first, for reasons already 
mentioned, and are followed by the polar regions. Besides these 
three primary categories there are the more localized climates of high 
altitudes (whose land vegetation, being largely comparable, we shall 
consider with that of the polar regions), of ' monsoon ' and ' Mediter- 
ranean ' types with warm and damp seasons alternating with dry 
ones, and of equable ' oceanic ' and widely-extreme ' continental ' 
types (see pp. 11-12). To the three primary climatic groupings the 
main vegetational belts of the world largely correspond, with local 
variations engendered by localized climatic and other features. 

The climates of temperate and adjacent lands are mostly fairly 
warm and moist, at least in the favourable periods. They exhibit 
rather marked seasonal and diurnal fluctuations, and also vary greatly 
from place to place. The mean of the warmest month each year 
is normally above io° C. (50 F.) and the annual precipitation is 


widely more than 762 mm. (30 inches). There is a marked difference 
between winter and summer light-climates and temperatures, and 
often, precipitation. The vegetation tends to be fairly luxuriant at 
least in favourable situations, with trees and shrubs widely dominat- 
ing but herbaceous plants usually exceeding them in number and 
variety. Most areas having a ' Mediterranean ' type of climate, 
with hot and dry summers but with other seasons that are damp 
and not too cold for plant growth, are included here, their vegetation 
being often dominated by leathery-leafed shrubs but including 
many bulbous and ephemeral herbs. The main vegetation-types of 
temperate and adjacent lands are dealt with in Chapter XII. 

The climates of polar lands and high altitudes are mostly rigorous, 
with the mean of the warmest month usually below io° C. Pre- 
cipitation is mostly in the form of snow and widely less than 254 mm. 
(10 inches) per annum, though owing to the prevailingly low tem- 
peratures the relative humidity may be high and the evaporating 
power low. There are wide seasonal fluctuations in most polar 
regions and wide diurnal ones in most alpine areas. In the higher 
latitudes there is continuous light in summer and darkness in 
winter. The vegetation is mostly low and scant — of dwarf shrubs, 
herbs (including many of grass habit), Lichens and Mosses. The 
main vegetational types of polar lands and high altitudes are dealt 
with in Chapter XIII. 

The climates of tropical and adjacent lands are warm and widely 
humid, with the mean temperature of the coldest month usually 
above i7'8° C. (64 F.) and the rainfall often heavy {e.g. 200-400 cm.). 
Frost and snow are usually unknown, the conditions being torrid 
and widely equable, with often little or no seasonal variation. The 
vegetation ranges from the world's most luxuriant rain forest to 
various scrub, grassland, and desert communities as the available 
water decreases. Most ' monsoon ' areas of alternating wet and dry 
seasons, commonly dominated by deciduous trees and shrubs which 
lose their leaves to conserve water during dry periods, are included 
here. The main vegetational types of tropical and adjacent lands 
are dealt with in Chapter XIV. 

It should be noted that the distinction between even ' oceanic ' 
(' maritime ', or ' insular ') and uneven ' continental ' climates is 
largely one of degree, being irrespective of latitude or temperature- 
relationships and consequently found in all of the above three primary 
climatic groupings. In general the oceanic extreme occurs on land 
where the prevailing winds come off the sea and are consequently 


moist and cloudy ; its areas tend to be- well vegetated, often with 
broad-leafed forests or verdant pastures. The continental extreme, 
on the other hand, is usually found far inland from the ocean and 
tends to have low relative humidity and precipitation, though 
exhibiting wide seasonal and daily fluctuations especially of tempera- 
ture. The summer here is commonly sunny and warm but dry, 
the winter being relatively cold, so that vegetation tends to be 
limited, often consisting of drought-resistant Grasses, Heaths, or 
desert plants. 

The Ideal Plant 

At this point will be given a brief account of the structure and 
adaptation of a multicellular ' higher ' plant, such as a member of 
the Angiosperms which top the * evolutionary tree ' and are dealt 
with at the end of the next chapter. Such flowering plants make 
up most of the bulk of modern vegetation, give us very many of 
our foods and other necessities of life, and consequently loom largest 
in our plant geographical and allied studies. 

Our ideal plant, as we may thus conceive it, will consist of (i) 
roots for anchoring in the ground and absorption from it of water 
and soluble nutrients, (2) a stem to hold the leaves and reproductive 
parts aloft, (3) green leaves to manufacture food substances in the 
light, and (4) flowers to produce seeds and so effect reproduction. 
Such features are too familiar to require illustration. 

Each main portion of a higher plant is composed of ' cells ', which 
are minute and often box-like structural units that are variously 
adapted to cover different needs. Cells of one kind are commonly 
aggregated together to form ' tissues ' of particular form and function. 
Thus some cells are for conduction — particularly of water and 
dissolved salts upwards from the roots and of elaborated materials 
downwards from the leaves — and are consequently elongated and 
often pipe-like. Other cells have greatly thickened walls and give 
tensile strength to roots and rigidity to aerial parts of the plant — 
especially in the latter instance when aggregates of them are situated 
near the periphery, as they commonly are in stems. Many cells on 
the other hand remain thin-walled and serve the purpose of aeration, 
food-storage, or mere ' packing ', while some may perform more 
than one function either concurrently or consecutively. All kinds 
of cells are produced from undifferentiated thin-walled ' meriste- 
matic ' ones which divide actively, for example in the growing-points 


(meristems) situated near the apices of stems and roots. Fig. 19 (B 
and C) shows stem-sections of higher plants with the main types 
of tissues and examples of their disposition. 

The above references are chiefly to the more or less solid walls of 
plant cells. But all these cells are alive, at least in youth — for they 
contain a viscous and very heterogeneous fluid known as protoplasm, 
which is the living matter of the plant. It is in the protoplasm that 
occur the extremely complex sequences of events which integrate 
into what we know as life, and which include the processes enabling 
the protoplasm to increase itself. This increase forms the basis of 
growth, which normally involves increase in size of the cell until 
it reaches a maximum and thereupon divides into two daughter 
cells. The daughters then repeat the process, and as a result of 
numerous repetitions of this activity the plant as a whole grows in 
size. Another activity going on in all living cells is the slow oxidative 
' burning ' known as respiration, which gives to living organisms the 
energy required for their life-processes. 

Besides the apical meristems by which plant organs grow in 
length, there is, in the stems and roots of many long-lived higher 
plants, a layer of actively dividing cells (the ' cambium ') which add 
daughters radially on either side and so lead to growth in girth. 
When this takes place year after year in regions of fluctuating 
climate, where cells of different sizes are produced at different 
seasons, annual ' growth-rings ' are formed which may easily be 
seen in most timbers. In addition there are meristems in buds 
whose behaviour — varying from dormancy to active elongation — ■ 
greatly affects the ultimate shape of plants. These and other growth 
phenomena are largely controlled by special chemical substances 
produced by the plant, and in ways which are only nowadays being 
elucidated. These plant growth substances, for example, may 
stimulate the elongation of cells in some tracts while inhibiting that of 
others — resulting in curvature of an organ in relation to a directional 
stimulus, such as light, which itself affects the production or 
availability of the chemical stimulant. Other substances inhibit 
growth, an example being produced by many terminal buds ; 
accordingly it is only when such inhibitors are removed that the 
lateral buds grow out actively (hence the sprouting of a hedge after 
clipping, and of pasturage after close grazing). For a general survey 
of this fascinating and important subject, see Professor L. J. Audus's 
Plant Growth Substances, second edition (Leonard Hill, London, 
pp. xxii + 553> 1959)- 



Plants cannot exist without water, though different kinds require 
it in very different amounts. Our ideal plant must be well adapted 
in its water economy to the prevailing conditions ; thus if water is 
scarce, it must have some means of keeping down the loss which 
takes place continuously from its aerial parts in the process known 
as transpiration. This economy may be effected by such devices 
as a thick and impervious bark or waxy or hairy covering, by pro- 
tection of the ' breathing pores ', or by reduction of the total surface. 
Often more than one method is employed by a plant, which at the 
same time will have to be adapted to other factors of the environment. 
Through long processes of evolution, involving among other things 
the elimination of unsuitable features, different kinds of plants have 
become adapted to different environments, and this, as we shall see 
for example in Chapter III, is one of the most fundamental bases 
of their distribution and consequently of plant geography. 

Plant Sociology 

Although opinions vary as to what constitutes a species (broadly 
speaking, a kind), we all have some conception of how similar 
individuals, whether plants or animals, make up such an entity. 
The numerous individuals comprising a particular species, while by 
no means all exactly identical, nevertheless are closely comparable 
in most respects, and normally have the appearance of belonging 
to the same kind. We have already observed that different plant 
species and other entities, often of many and various groups, become 
associated together in nature to compose what we term vegetation. 
This is made up of more or less definite plant communities, related 
at least in part to local conditions. Each of these communities is 
characterized by its own particular form (physiognomy), and in 
most cases also by one or more predominant species. 

Plant sociology, also called phytosociology, is, strictly speaking, 
the study of the plant communities that make up vegetation — 
including their inception and formation, their structure, and, above 
all, their composition. Accordingly some parts of this subject, and 
particularly the composition of plant communities, are of vital 
interest to the plant geographer, even as the distribution of these 
communities forms an important part of his study. But in spite 
of a wide overlap of material, students of the two disciplines 
approach their problems and subjects from different angles of 
interest, and so it is not proposed to consider plant sociology 


here, except in so far as it may help us to understand our own 
problems. 1 

The Animal Side 

As animals are so largely dependent upon plants for food, shelter, 
and other requisites of life, their geography and ecology tend to 
be less fundamental than those of plants, or at all events less closely 
related to the physical environment. Nevertheless the animal side 
of the picture of life (and in particular, Man's influence) must be 
vividly borne in mind by students of plant geography. Thus we 
shall see in Chapter IV how numerous plants depend upon animals, 
in many and various ways, for the dispersal of their seeds and 
fruits. Later on, in the chapters on vegetation-types, we shall be 
repeatedly reminded of how animals modify vegetation during their 
feeding and other activities, often favouring the growth, or very 
existence, of one species while discouraging that of another, and 
profoundly affecting the vegetation locally. In these and other ways, 
Man is apt to have the greatest influence of all. Many plants, 
too, depend on animals for pollination and hence fertilization of 
their flowers ; here, at least in the absence of vegetative means of 
propagation, reproduction will not normally take place without 
animal aid. All of these features can, and frequently do, affect 
the spreading and ultimate distribution of plant species. It is 
therefore not surprising that many areas, such as Australia and 
South Africa, have both a floristic and faunistic character and 
identity of their own, their (often peculiar) plants and animals going 
hand in hand, so to speak. Furthermore, animal geography often 
corroborates the conclusions of plant geography, and offers splendid 
evidence of evolutionary trends in its fossil record. It also appears 
to corroborate migrational tendencies in its suggestion of certain 
* land-bridges ' and ' refuges '. 

In view of the closeness with which the two are linked in nature, 
there is much to be said for the study, which has increased in 
popularity in recent decades, of plants and animals as they exist 
together in joint ' biotic ' communities. But whereas the particular 
physical conditions in an area are more or less vividly expressed in 
the local plant cover, which forms, as it were, a living framework, 

1 Interested readers are referred to the standard work on the subject by Dr. J 
Braun-Blanquet (Plant Sociology, McGraw-Hill, New York & London, pp. 
xviii -f- 439, 1932), or the second German edition (Pflanzensoziologie : Grundziige 
der Vegetationskunde, Springer, Wien, pp. xi + 631, 1951). 


it is only secondarily that this in turn largely conditions the animal 
population — which thus becomes a subordinate characteristic of the 
locality and is usually less evident and immediately significant than 
the vegetation, at least on land. Indeed, where there are no suitable 
plants there can be no animals living normally. As M. D. Haviland 
puts it in the work cited at the end of Chapter XVII, 

' It is the faithful correlation of plant growth with the physieal environ- 
ment, especially to the important factor, or complex of factors, called 
" climate ", that leads us naturally to define the main types of land 
environment in terms of plant life as Woodland, Grassland and Desert 
. . . for vegetation is the apparel of scenery. As Darwin wrote : " A 
traveller should be a botanist, for in all views plants form the chief 
embellishment." But when the zoologist, forsaking botanical terms, 
tries to classify environments in the language of his own science, he 
cannot construct a workable scheme ... he finds that he must fall 
back on the language of the botanist or geologist.' 

In general, zoologists have not been very successful in recognizing 
definite animal communities of a complex nature, and their study 
in individual species of adaptive response to particular environments 
tends to be of less immediate significance than that of botanists 
with plants. Thus, whereas the marked dwarfing of many arctic 
and alpine plants is related directly to exposure to harsh physical 
environments, many similarly striking animal characteristics, such 
as broad teeth for grinding seeds and special organs for climbing 
trees, are related to the climatic conditions only indirectly through 
plant response. Nevertheless, as pointed out by Professor G. E. 
Hutchinson (in litt.) y there are a number of known cases of warm- 
blooded animals responding directly to climate. For example, 
boreal Mammals not only have under-fur but also are of larger 
absolute size, and have shorter ears and tails, than their southern 
counterparts, while almost all desert Mammals and Birds are pale 
even if nocturnal, and insular races of Birds have relatively large 
beaks and feet. But in spite of such exceptions, and others which 
act in the opposite direction (such as the striking adaptations of 
many flowers to insects in relation to pollination), plant response 
to climate is usually direct whereas that of animals tends to be 
indirect. Then again, animals are usually mobile, and individuals 
may wander or migrate vast distances. Consequently, apart from 
such connections as those mentioned above, animals tend to be of 
less geographical significance than plants, in the sense that they do 


not characterize areas to the same extent, and for our present purpose 
seem best considered as a mere factor of the environment. 

This recognition of the more fundamental role of plants does 
not seem to be weakened by the realization that, often, plants and 
animals have evolved together and are necessary for one another's 
existence. For even in the case of flesh-eating animals, sooner or 
later, as we trace back the food-chain, we come to the ultimate point 
of dependence upon green plants. Furthermore, the animal geo- 
grapher is not necessarily of much help to us, for the areas and 
boundaries which he recognizes (e.g. Fig. 2, A) are often very different 
from ours (e.g. Fig. 2, B), and he is prone to take for granted that 
the vegetation (which he considers simply as part of the environ- 
ment) is a mere response to local conditions. Yet a plant community, 
quite apart from its historical implications, gives us many clues to 
the nature of the environment because its component members 
exhibit recognizable responses to physical features. No such general 
virtue is displayed by animal communities, if indeed these can be 
satisfactorily recognized. 

Recent books on animal geography, with useful bibliographies 
suggesting further reading, include R. Hesse, W. C. Allee, & K. P. 
Schmidt's Ecological Animal Geography, second edition (Wiley, 
New York, pp. xiii -f 715, 1951), F. L. de Beaufort's Zoogeography 
of the Land and Inland Waters (Sidgwick & Jackson, London, pp. 
viii + 208, 195 1), Sven Ekman's Zoogeography of the Sea, translated 
by Elizabeth Palmer (Sidgwick & Jackson, London, pp. xiv + 417, 
1953), and Philip J. Darlington's Zoogeography : the Geographical 
Distribution of Animals (Wiley, New York, pp. xiii + 675, l 9Sl)- 

Some Earlier Works on Plant Geography 

In English : 

Anonymous and other early works include The Geography of Plants (The 
Religious Tract Society, London, pp. vi -f- 7-192, undated), J. 
Barton's A Lecture on the Geography of Plants (Harvey & Darton, 
London, pp. 1-94 and index, etc., 1827), an d R- B. Hinds's The 
Regions of Vegetation ; being an analysis of the distribution of vegetable 
forms over the surface of the globe in connection with climate and 
physical agents (Palmer, London, pp. 1-140, 1843). 

Meyen, F. J. F. (1846) : Outlines of the Geography of Plants : with 
particular enquiries concerning the native country, the culture, and the 
uses of the principal cultivated plants on which the prosperity of nations 



u <u 
bC o 

S 6 

a ° 

o 2 

C bo 







iV based, translated by M. Johnston (Ray Society, London, pp. 
x + 422). Follows the German edition', published in Berlin in 1836. 
Of historical interest as indicating the teachings of the day, including 
Man's dependence upon plants, but with over-emphasis on latitude 
as the limiting factor in plant distribution. 

Daubeny, Charles, ed. (1855) : Popular Geography of Plants ; or, a 
botanical excursion around the world (Lovell Reeve, London, pp. 
xl + 370)- A very readable illustrated account following Meyen's 
arrangement and, although often unreliable, of historical interest as 
indicating the type of work apparently favoured by the intelligent 
layman of a century ago : by ' E.M.C, with a well-written and 
penetrating preface by its eminent editor. 

Pickering, Charles (1876) : The Geographical Distribution of Animals 
and Plants. Part II. Plants in Their Wild State (Naturalists' 
Agency, Salem, Mass., pp. 1-524 and additional maps). A sumptuous 
but evidently rare publication of some interest and foresight. 

Schimper, A. F. W. (1903) : Plant-geography upon a Physiological Basis, 
translated by W. R. Fisher, revised and edited by Percy Groom and 
I. B. Balfour (Clarendon Press, Oxford, pp. xxx 4- 839 and 4 
additional maps). Still the great reference book on the subject in 
English, profusely illustrated and a commendable feat for its time, 
though in places unreliable and now largely outdated. A detailed 
up-to-date work in English, planned along modern lines and executed 
in the light of the latest knowledge, is badly needed to supersede it. 

Warming, E., et. al. (1909) : Oecology of Plants (Clarendon Press, Oxford, 
pp. xi 4- 422). For many years a standard source-book mainly on 
the ecological side but also describing the main vegetation-types. 

Hardy, M. E. (1913) : A Junior Plant Geography (Clarendon Press, 
Oxford, pp. 1-192). A light-weight but useful discourse on some 
aspects of the subject — like the next work, illustrated and readable 
though outdated and not always reliable. Some copies have been 
seen entitled ' An Introduction to Plant Geography '. 

Hardy, M. E. (1920) : The Geography of Plants (Clarendon Press, 
Oxford, pp. xii 4- 327). Consists chiefly of a discursive account of 
the more obvious vegetational features of the main land-masses and 
climatic regions (cf. above), largely ignoring aquatic habitats. 
Reprinted up to 1952. 

Campbell, D. H. (1926) : An Outline of Plant Geography (Macmillan, 
London [and New York], pp. ix 4- 392). Illustrated and readable : 
concerned chiefly with climatic zones and areas and their land flora 
and vegetational characteristics, but loosely written and frequently 
inaccurate, and omitting many topics which might with advantage 
have been treated. 

Newbigin, M. I. (1936) : Plant and Animal Geography (Methuen, 
London, pp. xv 4- 298). See also the practically identical ' second 


edition ' (Dutton, New York, pp. xv -f- 298, 1948). A stimulating 
and usually reliable outline of many aspects of biogeography. The 
so-called second edition, although recent, is practically a reprint that, 
unfortunately, fails to remedy some misconceptions and to correct 
errors particularly in those chapters for which the original author 
was not responsible. 

Wulff, E. V. (1943) : An Introduction to Historical Plant Geography, 
translated by E. Brissenden (Chronica Botanica, Waltham, Mass., 
pp. xv -f- 223). Useful in elucidating the origin and development 
of floras as opposed to their composition, ecology, and other aspects. 
A succeeding volume was published later in Russia (see below). 

Cain, S. A. (1944) : Foundations of Plant Geography (Harper, New York 
& London, pp. xiv -j- 556). An important though rather technical 
survey of the history and interpretation of many phenomena of 
vascular plant distribution on land. Does not deal with aquatic 
habitats or lower groups of plants, and is professedly not a descriptive 
plant geography. 

Croizat, L. (1952) : Manual of Phytogeography (Junk, The Hague, pp. 
viii + 587 and 106 additional illustrations). Considers plant geo- 
graphy simply the study of plant dispersal, being ' that branch of 
botany which integrates plant-migrations in time and space '. A 
large part (pp. 68-399) is occupied by treatment of the ' intercon- 
tinental dispersal ' of various (mainly tropical) Angiosperms. Often 
opinionated and sometimes crotchety : nor is the coverage in accord- 
ance with the subtitle which claims the work to be ' an account of 
plant-dispersal throughout the world '. 

Good, Ronald (1953) : The Geography of the Flowering Plants, second 
edition (Longmans, London etc., pp. xiv + 452). An illustrated 
manual of the distributions of flowering plants and the factors 
controlling them ; does not deal with vegetation or with lower 
plants. Nevertheless a valuable work, and widely considered the 
standard one on the floristic side of the subject. The second edition 
should be used rather than the first, which was published in 1947. 

Turrill, W. B. (1953) : Pioneer Plant Geography : the phytogeographical 
researches of Sir Joseph Dalton Hooker (NijhofT, The Hague, pp. 
xii -f- 267). Readable and instructive, with up-to-date comments, 
as well as historically interesting. 

In other languages : 

Humboldt, A., & A. Bonpland (1805) : Essai sur la Geographic des 
Plantes ; accompagne d'un tableau physique des regions equinoxiales 
(Paris, pp. i-xii -f- 13-155 and map). One of the main foundations 
of our subject, followed by a German edition in 1807, and also by 
kindred works in various languages ; of great historical interest. 

Schouw, J. F. (1822) : Grundtraek til en almindelig Plant egeographie 


(Kjobenhavn, pp. x -f- 463 and 4 additional illustrations). Fol- 
lowed the next year by a German edition entitled Grundziige einer 
allgemeinen Pflanzengeographie (Berlin, pp. xix + 528 and 4 additional 
illustrations). Of historical interest. 

Rudolph, L. (1853) : Die Pflanzendecke der Erde (Berlin, pp. xiv -j- 416 
and additional plates ; also ' Supplementheft ' of 34 pages, published 
in Berlin in 1859). An early semi-popular treatment. 

DeCandolle, Alphonse (1855) : Geographic Botanique Raisonnee ou 
exposition des {aits principaux et des lois concernant la distribution 
geographique des plantes de Vepoque actuelle (Paris & Geneve, vol. I, 
pp. xxxii + 606 and additional maps, and vol. II, pp. 607-1366). 
An early detailed synthesis of much of our subject, now chiefly of 
historical interest. 

Grisebach, A. (1877-8) : La Vegetation du Globe d'apres la disposition 
suivant les climats . . ., translated by P. de Tchihatchef (Bailliere, 
Paris, vol. I, pp. xvi -f- 765 and additional map, 1877, an d v °l- H> 
pp. vi -f 905, 1878). Discursive but interesting, at least historically. 

Engler, A. (1879-82) : Versuch einer Entwicklungsgeschichte der Pflanzen- 
welt, insbesondere der Florengebiete seit der Tertidrperiode (Engelmann, 
Leipzig, vol. I, pp. xviii + 202 and additional map, vol. II, pp. 
xiv + 386 and additional map). Includes suggested explanations of 
plant distribution but was soon in part superseded. Vol. I, published 
in 1879, deals with the extratropical regions of the northern hemi- 
sphere, and vol. II, published in 1882, with the tropical regions and 
the remainder of the southern hemisphere. 

Grisebach, A. (1880) : Gesammelte Abhandlungen und kleinere Schriften 
zur Pflanzengeographie (Engelmann, Leipzig, pp. vii -f- 628). Various 
contributions including the general and lengthy ' Berichte iiber die 
Fortschritte in der Geographie der Pflanzen ' (pp. 335-556). 

Contejean, C. (1881) : Geographie Botanique ; influence du terrain sur 
la vegetation (Bailliere, Paris, pp. 1-144). Deals chiefly with habitat 
differences and their effect on local flora. 

Goeze, E. (1882) : Pflanzengeographie fiir Gartner und Freunde des 
Gartenbaues (Ulmer, Stuttgart, pp. xiv + 476). A general treat- 
ment, primarily but by no means exclusively for horticulturists. 

Drude, O. (1890) : Handbuch der Pflanzengeographie (Engelhorn, Stutt- 
gart, pp. xvi + 582 and additional map). A general manual, illustrated 
chiefly by maps, of the subject as then developed, by a renowned 
investigator of the time. Followed in 1897 by a French edition 
entitled Manuel de Geographie Botanique (Klincksieck, Paris, pp. 
xxiii + 552). 

Solms-Laubach, H. zu (1905) : Die leitenden Gesichtspunkte einer 
allgemeinen Pflanzengeographie (Felix, Leipzig, pp. ix + 243). A 
briefer, unillustrated account of much of the subject, with some 
novel ideas. 


Graebner, P. (1910) : Lehrbuch der allgemeinen Pflanzengeographie nach 
entzcicklungsgeschichtlichen und physiologisch-okologischen Gesichts- 
punkten mit Beitrdgen von Paul Ascherson (Quelle & Meyer, Leipzig, 
pp. viii + 303)- A general, illustrated account of the history and 
composition of various floras and vegetation-types, paying due regard 
to their ecological bases. 

Hayek, A. (1926) : Allgemeine Pflanzengeographie (Borntraeger, Berlin, 
pp. viii + 409 and 2 additional maps). A more modern account 
along similar general lines to the last, but illustrated by only a very 
few diagrams and maps. 

Diels, L. (1929) : Pflanzengeographie, third edition (Grunter, Berlin & 
Leipzig, pp. 1-159 and additional map). Useful as presenting a 
largely modern account in outline of many aspects of the subject 
in handy pocket form. Later editions (not seen) have since appeared. 

Rubel, E. (1930) : Pflanzengesellschaften der Erde (Huber, Bern-Berlin, 
pp. viii + 464 and map). Describes and illustrates the main plant 
communities and vegetation-types of the world. 

Warming, E., & P. Graebner (1933) : Lehrbuch der okologischen Pflanzen- 
geographie, fourth edition (Borntraeger, Berlin, pp. viii -f- 1158). A 
valuable illustrated work dealing particularly with ecological and 
vegetational aspects. 

Schimper, A. F. W. (1935) : Pflanzengeographie auf physiologischer 
Grundlage, * third ' edition, revised by F. C. von Faber (Fischer, Jena, 
vol. I, pp. xx -{- 588, and vol. II, pp. xvi + 589-1613 and 3 additional 
maps). An illustrated, extensive work weighing about 10 lb. and 
dealing mainly with the natural flora and vegetation of different 
zones and regions, after some consideration of ecological factors and 
principles. Generally reliable in those aspects of our subject with 
which it deals. Although called (in German) the third edition, this 
was in reality the second, as the so-called second edition was 
merely a reprint of the first edition. 

Alekhin, V. V. (1944) : [Geography of Plants], in Russian only (State 
Publisher, Moscow, pp. 1-455 and 2 maps). An illustrated account 
of many of the distributional and ecological aspects of the subject. 

Wulff, E. V. (1944) : [Historical Plant Geography : history of the floras 
of the world], in Russian only (Akademiya Nauk SSSR, Moscow- 
Leningrad, pp. xix -j- 546). This is the second volume, mainly on 
the origins of the floras of the different regions of the world, of a 
projected three-volume work of which the first was translated into 
English (see above) and the third was apparently never completed, 
the author being killed in 1941 during the siege of Leningrad. 

Gaussen, H. : Geographic des Plantes, second edition (Colin, Paris, pp. 
1-224, z 954)- Gives a brief but useful account of many aspects of 
the subject. 

Chapter II 


Classification and Nomenclature 

To enable us to name and deal effectively with the almost infinite 
variety of plants inhabiting the world, it is necessary to sort into 
groups those which seem to have the closest affinity or, at least, 
the greatest outward similarity. These groups in turn have to be 
aggregated into larger groupings, and so on, to create a hierarchical 
system of classification which we also like to think bears a close 
relation to evolutionary history. Thus the members of a group 
which look closely alike probably bear a ' blood relationship ' in 
being descended from a common ancestor at no very remote period 
of geological time ; indeed in some instances such a relationship 
has been experimentally demonstrated. Biologists, and this includes 
botanists, may in some cases disagree about the definitions, names, 
and limits of this hierarchy of groups, but for general purposes 
(and in descending order, from large to small) these groupings 1 may 
be listed as follows : 

Divisions or phyla (sing, phylum) : the major (highest) groupings 
used in classifying plants, with names normally ending in -phyta, 
those commonly recognized being the Schizophyta, Thallophyta, 
Bryophyta, Pteridophyta, and Spermatophyta, and each consisting 
of one or more 

Classes : the next commonly recognized units, plentifully 
exemplified below, and each consisting of one or more 

Orders : each of which has its name ending in -ales, and in turn 
consists of one or more 

Families : these, except in a few long-established instances, have 
their names ending in -aceae. Usually the members of a family all 
have some recognizable characteristic or characteristics, some com- 
mon ' stamp ' ; they are grouped into one or more 

1 Also called taxa (singular taxori), regardless of rank. 


Genera (sing, genus) : the members of each genus usually look 
alike in a number of features and constitute one or more 

Species : these represent the smallest unit of classification in 
general use, being those whose members show a broad similarity. 
Biologists differ in their conception of what constitutes a species, and 
indeed the term is scarcely capable of satisfactory definition. How- 
ever, for the great majority of animals and many plants a species is, 
roughly speaking, constituted by all those individuals which are able 
to interbreed among themselves but are unable to breed, at least at 
all freely, with members of other groups. The individuals of a 
species are by no means identical but form a more or less variable 
population in which some entities are often recognizable as subspecies 
(written subsp. or ssp.), or as still more subordinate varieties (written 
var.) or formae (written f.). 

Even as species are divisible into subspecies, so are the major 
groups often divided into subphyla, subclasses, etc. The individual 
is the ultimate unit, but inasmuch as no two individual plants can 
be exactly identical, any more than two individual persons can be, 
the smallest recognizable unit of classification is the biotype, con- 
sisting of all those individuals which have the same genetical make-up. 
Thus most species consist of a large number of biotypes that differ 
slightly in their inheritance. 

The scientific name of each species is normally made up of two 
Latin or latinized words of which the first is the name of the genus 
to which it belongs and the second is its own specific epithet, usually 
having some descriptive or historical connotation. The initial letter 
of the first, or generic, name is always capitalized, that of the specific 
epithet nowadays being customarily left ' small '. Unfortunately, 
English or other ' popular ' names are too unreliable to employ at 
all widely, particularly because the same name is apt to be used for 
different plants in different places or by different people, and also 
because it is undesirable to have the same plant known under different 
names in different places or sometimes even in the same place. 
Moreover, it is confusing to have more than one combination of 
names in use for a single entity, and so the Latin one is agreed upon 
and employed internationally by scientists. 

We will now consider briefly what seem for our purpose to be 
the main classes of the plant kingdom, each being treated under the 
primary heading of the phylum (division) to which it belongs. The 
sequence followed is probably indicative of evolutionary history in 
broad outline. After a brief general account of the characters of 


each of our chosen classes, we will deal with its modes of nutrition 
and reproduction and also, in the broadest terms, with its main 
habitats, distribution, and importance — both economically and as 
a component of natural vegetation. 


Bacteria : These are very simple, exceedingly minute, and 
virtually ubiquitous organisms. Most types consist of single 
spherical, rod-shaped, branched, or variously curved cells, with or 
without delicate superficial thread-like processes known as flagella 
{sing, flagellum), through the action of which they may attain a fair 



Fig. 3. — Various types of Bacteria. Among those causing serious human diseases 

are B, anthrax; H, typhoid fever; N, cholera; Q, tuberculosis; R, leprosy; 

S, diphtheria; T, meningitis ; U, pneumonia; V, dysentery; X, tetanus. (Mostly 

X 1000, but Q and V considerably more magnified.) 

degree of motility in liquid media. In other types the cells may 
adhere together in small groups or chains, or remain attached by 
the ends to form regular filaments. Various types of Bacteria are 
shown in Fig. 3, most of them being magnified about 1,000 times. 
Bacteria are found in vast numbers almost everywhere : in soils 
and the atmosphere, in fresh and salt waters, and in many of the 
most unlikely and unsavoury ' habitats '. Normal soils with a fair 
percentage of organic matter contain on the average from 2,000,000 
to 200,000,000 Bacteria per gram, while manure and sewage may 
contain greater numbers still. 

Bacteria multiply principally by simple cell division, sometimes 


as frequently as every twenty minutes — but only for a time, the 
chief limitation being their food supply. They may also form 
spores which in some cases are highly resistant. Recently, convinc- 
ing evidence of a sexual process has been obtained in some Bacteria, 
but it is not known how frequent such a process may be in nature. 

In their modes of nutrition Bacteria vary greatly : for although 
commonly they are either (1) saprophytic, deriving their energy and 
materials for life and growth from dead and usually decaying organic 
matter, or (2) parasitic, depending similarly on living organisms; 
there are also (3) many forms which can build up their bodies and 
live from carbon dioxide obtained from the air or water and energy 
liberated in the oxidation of inorganic compounds or even elements. 
Such organisms are said to be chemosynihetic, and examples of the 
substances oxidized by them are sulphur, hydrogen sulphide, nitrites, 
ammonia, hydrogen, and, apparently, iron and manganese com- 
pounds. The members of one interesting group of sulphur-oxidizing 
Bacteria, known as the Purple Bacteria, contain pigments enabling 
them to absorb radiant energy from light, and they appear to practise 
some kind of photosynthetic process which may be a prototype of that 
occurring in ' normal ' green plants. Other types that seem properly 
referable to the Bacteria are green, through the inclusion of chloro- 
phyll of a kind. But it seems improbable that the earliest living 
organisms possessed real chlorophyll or obtained their energy 
through such an elaborate series of reactions as are involved in 
photosynthesis (cf. p. 32). Rather is it considered likely that some 
of these peculiar Bacteria indicate means by which elementary 
organisms obtained their energy and other requisites of life before 
either chlorophyll or any form of photosynthesis was evolved. 
Consequently it seems most reasonable to start our sequence with 
this group. 

Certain Bacteria are of major importance in causing diseases— 
particularly of animals, and including some of those most deadly 
to Man — while various other Bacteria cause the decay and breakdown 
of dead matter, or make available food-substances for higher plants, 
or produce chemical ions of many kinds. With their infinitesimal 
size and often resistant spores, they are among the most widespread 
of living organisms, being carried by air or water currents or in the 
bodies of animals practically everywhere in the world and its sur- 
rounding atmosphere. Nevertheless, as they are so minute, they 
play only a very minor direct role as components of most types of 
vegetation. Exceptions are afforded by some aquatic muds, in which 




Bacteria may dominate, and also, of course, in diseased or decaying 
systems. They may also be the pioneers in the colonization of some 
bare areas. 

Cyanophyceae : These are the so-called ' Blue-green Algae ', or 
Schizophyceae, but they seem to have their closest relationship with 
the Bacteria, which they resemble particularly in their lack of a 


Fig. 4. — Various Blue-green Algae (Cyanophyceae). A, Rivularia ( X about 600) • 
B, Aphonothece (X about 1500); C, Merismopedia ( X about 200); D, Oscillatoria 
(X about 800); E, a species of Nostoc, consisting of filaments embedded in a 
gelatinous matrix (X about 300); F, Gloeocapsa: a single-celled individual and 
colonies of two, three, and four cells (X 1285). 

typically organized nucleus in the cell, in their method of cell 
division, and in the obscurity at all events of any sexual reproduction. 
They consist either of single cells, of cells joined end-to-end to form 


filaments, or of colonies in which either individual cells or filaments 
are held together in gelatinous masses. Examples, variously 
magnified, are shown in Fig. 4 ; but whereas the individual cells 
are microscopic and often exceedingly minute, the colonies may be 
many centimetres in diameter and of considerable bulk. 

The Cyanophyceae contain chlorophyll of a sort and mostly live 
by photosynthesis although some appear to be at least partly sapro- 
phytic. The chlorophyll is diffused through the outer layers of 
protoplasm instead of being accumulated in special bodies as it is 
in higher plants. Cyanophyceae also contain bluish and/or reddish 
pigments, the former of which, with the chlorophyll, gives to many 
a blue-green coloration — hence their popular name. Others, how- 
ever, are very differently coloured. Multiplication is by simple cell 
division or the formation of spores. The filaments or parts of 
filaments of many exhibit motions of various but characteristic kinds, 
including glidings and oscillations, the mechanisms of which are 
not understood. 

Cyanophyceae are very widespread, occurring especially in a range 
of freshwater and damp to marshy habitats. However, though often 
abundant, they make relatively little showing as components of 
vegetation — except sometimes in lakes or ponds, where they may 
form a ' bloom ', or on stones in fairly rapid streams where they 
provide a mucilaginous mat, in which other organisms such as 
Diatoms thrive. They tend to be important in arctic regions where 
they often form dark colonies on damp soil, in marshes, and about 
the margins of tarns. Ecologically they may be of some importance 
as the initial colonists (pioneers) on bare rock and other surfaces. 
They have little economic significance except as nuisances in fouling 
water supplies — to which they may impart a disagreeable odour and 
taste, sometimes killing fish or even cattle. 


Chlorophyceae : These are a large and diverse group of Algae 
(seaweeds, aquatic ' slimes ', etc.) having chlorophyll located in well- 
defined bodies called chloroplasts and not normally masked by other 
pigments. Consequently the plants are usually green in colour and 
the group is called the Green Algae. Characteristically they have 
cellulose cell-walls and store food in the form of starch. What 
appear to be the more primitive forms are microscopic, unicellular, 
and either motile by flagella or non-motile, the cells occurring singly 



\ /I 




A, Pleurococcus — unicel- 
lular, non-motile ( X 2470) ; B, Chlamydomonas — unicellular, motile ( X about 
500); C, Pleodorina, colonial, motile ( X 250); D, cells from a filament of Ulothrix 
(X 462); E, Enteromorpha intestinalis , a relative of Ulva (see G) (x i); F, Chara, 
a highly organized Stonewort (x ,1); G, Ulva lactuca, a Sea-lettuce (x about f); 
H, various Desmids ( X 220). 

or grouped into colonies. Advanced types are commonly attached to 
some object and consist of filaments or more substantial branched 
structures, or have the form of a flattened thallus (simple vegetative 
body lacking differentiation into true root, stem, and leaf) that may 
be several inches in diameter. Fig. 5 shows a range of different 
Chlorophyceae ; they are not merely extremely diverse but also 
appear to have undergone evolution along a number of different lines, 
most of which have proved to be ' dead ends '. Here are included 
the Desmids, which are freshwater forms consisting usually of a 
single cell that is sharply marked off into two symmetrical (and often 


complicatedly lobed) halves by a constriction around the centre. 
Also included are a wide range of freshwater and marine slimes and 
scums, the sometimes bulky Sea-lettuces, and, according to most 
students, the peculiar and fir-like Stoneworts. 

Different kinds of Green Algae are to be found in a vast array of 
habitats including damp soil, the surfaces of rocks, and the bark of 
trees. Here and on such objects as posts and palings they may 
form a green investment most typically on the pole-facing side, 
which tends to be less dried by the sun. But mainly they are 
aquatic, being especially abundant in freshwater lakes and streams, 
although numerous types occur in the sea. Many Green Algae are 
among the most widespread of plants, the group as a whole being 
virtually ubiquitous. Their nutrition is mainly by photosynthesis 
— that fundamental series of reactions on which they and practically 
all other forms of life depend, directly or indirectly. In this vital 
process, chlorophyll absorbs radiant energy from light and catalyzes 
the building up of simple materials into complicated carbohydrates 
in which the energy is locked — there to remain stored until it is 
liberated, for example by burning or the slower process of respira- 
tion. Such carbohydrates made by green plants constitute the main 
basic food materials of the world. Although some simulate higher 
plants, Algae do not need (or have) roots or other special absorbing 
organs, but take in the raw materials they require (chiefly water 
with certain salts and gases in solution) more or less all over their 

The Green Algae reproduce by various methods, several of which 
may be practised by the selfsame species. The chief types are 
asexual reproduction by cell division, fragmentation of the thallus, 
or the liberation of spores — which may swim actively by means of 
flagella. Or there may be sexual reproduction following conjuga- 
tion or, more often, the fusion of special bodies (gametes) which are 
frequently differentiated into large female and small male ones. In 
some cases both of the gametes and in others only the male ones 
are motile. In spite of their great diversity and virtual ubiquity, 
the Green Algae are of rather little importance except as food 
for aquatic animals ; but they are of great interest in indicating 
some of the lines along which evolution to higher plants may have 
taken place. They are widely important constituents of aquatic 
vegetation and are often dominant in freshwater pools, though as 
constituents of human or domestic animals' food they are at best 


Bacillariophyceae : These are the Diatoms, familiar to all 
microscopists, and comprise a large and important class of Algae that 
are all unicellular and microscopic, occurring singly or attached in 
filaments or chains, or grouped into colonies. The form is extremely 
various, as may be seen from Fig. 6, 
which shows a range of different types. 
The cell-wall is composed of two 
' valves ', overlapping one another like 
the halves of a pill-box, and is impreg- 
nated with silica. Its surface is finely 
and often beautifully sculptured, with 
extraordinary regularity and precision. 
The chloroplasts contain a brown pig- 
ment in addition to the all-important 
chlorophyll, and accordingly the colour 
of Diatoms both individually and en 
masse is usually a shade of brown or 

The nutrition of Diatoms is primarily 

Fig. 6.- 


(Bacillariophyceae). A, various forms (variously magnified) 
a colonial type, Licmophora flabellata ( X 70). 

by photosynthesis, food being stored in the form of oil. Their 
reproduction is normally by cell division, though occasionally sexual 
conjugation takes place, followed by the production of special 
' auxospores ', or these latter may be produced apomictically ; 
alternatively, small flagellated gametes or spores may*, be formed. 
Diatoms are abundant in both marine and fresh waters in practically 
all climates, forming a significant, and often the main, constituent 



of the plankton — the more or less passively floating or drifting plant 
and animal population of seas and lakes — and as such are of vital 
importance as the ultimate source of food of many fishes and other 
sea and freshwater animals. They are virtually world-wide in dis- 
tribution, being found, for example, on and in damp soil and upon 
as well as under the sea-ice even about the North Pole. When 
they decompose or are digested by animals, their siliceous valves 
usually do not decay but sink in considerable quantities to the 
bottom of the body of water, often forming extensive deposits of 
' diatomaceous earth '. This is widely used for scouring, filtering, 
insulation, and other purposes. 

Dinophyceae : These are the Peridinians or Dinoflagellates — 
usually motile, microscopic unicellular organisms of yellowish or 
brownish colour and sometimes of marked luminescence. A few 
are naked but the vast majority have cellulose walls, which are 
often composed of several sculptured plates. Fig. 7 shows three 
flagellated, motile examples ; but even those types which are non- 
motile and filamentous reproduce by motile spores (zoospores) that 
have the form of typical Dinophyceae. These 
have a particularly characteristic feature — two 
grooves at right-angles, one of which encircles 

Fig. 7. — Dinoflagellates (Dinophyceae). A, Gymuodinium, a type without plates 

( >C about 1560); B, Goniaulax, a type armoured with plates (x about 1200); 

C, Geratium, with plates and form-resistance ( X about 300). 


the cell transversely while the other runs longitudinally along 
one side. Two flagella are inserted where the grooves cross each 
other — an undulating one which lies in the transverse groove and 
appears to be largely responsible for the rotation of the organism, 
and a more normal looking one running down the posterior portion 
of the longitudinal groove and effecting movement forward. 

Nutrition is mainly by photosynthesis, food being stored as either 
starch or oil. Not only is a reddish eye-spot frequently present, 
but some types have colourless bodies and are saprophytic, while a 
few, at least, ingest solid food and so are animal-like in their feeding. 
As these are among the organisms that are on the border-line between 
animals and plants, they are liable to be claimed also by zoologists. 
Reproduction is chiefly effected asexually by the division of an 
individual into two dissimilar halves (the cells are often markedly 
asymmetric at first), after which each half regenerates the missing half. 

Peridinians are very widely distributed in both fresh and salt 
waters and may be especially abundant in the ocean. Thus in 
arctic seas they tend at some times of the year to outnumber even 
the Diatoms and, temporarily, to form the main constituent of the 
plankton. Consequently they are an important source of food for 
marine animals — including, ultimately, Fishes, Seals, and even the 
greatest Whales. 

Phaeophyceae : This large group, commonly called the Brown 
Algae or Brown Seaweeds, are characterized by their brown or 
olive-green colour which is due to the chloroplasts containing a 
special brown pigment in addition to chlorophyll. They are practic- 
ally all marine, being among the most abundant and familiar sea- 
weeds of temperate and more austral (southern) as well as boreal 
(northern) coasts. The thallus is multicellular and usually attached, 
but shows a very wide range of different forms, some examples 
of which are shown in Fig. 8. Though sometimes slender and 
filamentous, the thallus is more often complex. Frequently it is 
relatively massive, being differentiated into a disk- or root-like organ 
of attachment to tidal rocks or sea-bed objects, and a stem-like part 
of varying length and thickness bearing a ribbon- or leaf-like portion. 
This last may be branched or unbranched and is usually elongated 
and flexible, streaming easily with the current or, in shallow water, 
often floating. Such buoyancy is commonly aided by the inclusion 
of air bladders, which are usually conspicuous and large enough to 
' pop ' when trodden upon — as in the familiar Bladder Wrack {Fucus 





Fig. 8. — Some Brown Algae (Phaeophyceae). A, Ectocarpus, a filamentous type 

(X 45); B, Dictyota (x i); C, Fucus (X {); D, Agarum (X ?r); E, Chorda 

X i); F, Alaria (X ^o); G, Ulopteryx (X iV); H, Sargassum ( < |). 

resiculosus). Types of Brown Algae living between tide-marks, as 
many do, are usually whippy and tough enough to remain uninjured 
by the waves. In some of the giant Kelps the ' fronds ' have a 
relatively complex internal structure and may be around 200 feet 
in length. (Reports of much greater lengths do not appear to have 
been authentic.) 

The Brown Algae obtain their food for body-building, growth, 
and energy by means of photosynthesis, the raw materials for this, 
carbon dioxide and water, being absorbed over the entire surface, 
as are also the needed salts, etc., in solution. Reproduction is 
effected asexually by fragmentation of the thallus or by liberation 
of motile spores (zoospores), or, sexually, by the fusion either of 
two similar motile gametes or of dissimilar gametes. When there 
are dissimilar gametes one, the male, is motile and small while the 
other, the female, is non-motile and relatively large. Thus sexuality 


may here be comparable in several ways with that of most animals. 
Many types of Brown Algae have an alternation of sexual and asexual 
generations — usually of strikingly different sizes and forms, as in 
the case of Pteridophytes (see pp. 55 et seq.). 

Although a few simple types occur in fresh water, the vast majority 
of Brown Algae live in salt seas or in brackish lagoons and estuaries. 
They are abundant in the tropics, but tend to be still more prominent 
in colder waters, even persisting to the northernmost arctic shores. 
Being often of considerable size, they are commonly the most 
conspicuous features of many northern rocky shores between tide- 
marks and for some distance below, forming extensive and often 
almost pure ' beds '. These may cover and obscure the rocks or 
boulders, to which the plants are attached by holdfasts so tough that 
they are normally detached only during severe storms. Thereafter 
they may be cast up in large piles upon the beach, or float until 
they die and ultimately disintegrate. A few special kinds of Brown 
Algae, however, seem able to live indefinitely in a detached floating 
state, forming extensive masses ; of these masses the largest and 
most notable characterizes the Sargasso Sea in the western Atlantic 

The distribution of the Brown Algae seems to be world-wide 
wherever suitable sea-shores are found. Nor is their economic 
significance negligible. Thus some are extensively harvested for 
human food or animal fodder, particularly in eastern Asia, while 
their use as manure after being cast up during storms is widespread. 
The ash obtained by burning certain Kelps and Wracks, particularly, 
is still in some places an important source of iodine and potassium. 
Finally, owing to their peculiar food-storing and other biochemical 
activities, Brown Algae are nowadays an important source of often 
unique organic chemicals. An example of these is ' algin ', the 
production of which runs into about a thousand tons annually in 
the United States alone. 

Rhodophyceae : These are the Red Algae, or Red Seaweeds, 
which tend to be more prolific in different species but are generally 
less bulky and abundant as individuals than the Brown Algae. This 
is especially the case in the seas of temperate and boreal regions, 
where Red Algae may be little in evidence. They are characteristic- 
ally red or purplish owing to the presence in the chloroplasts of 
special pigments besides chlorophyll, although some may be greenish, 
bluish, olive, or brown. The thallus is again very various in form 


in the different genera or even species, ranging from filamentous 
(very rarely unicellular) to densely branched and coral-like (owing 
to encrustation with ' lime '), and from a flat blackish disk to a ribbon- 
shaped or widely expanded ' frond '. Attachment to the substratum 
is by disk-like holdfasts or special filaments ; or the whole plant 
body may form a close investment. Many Red Algae are very 
intricate and beautiful in form : a range of examples is shown in 
Fig. 9. The internal structure is peculiar and relatively complex, 
though none of the Red Algae approaches in size the larger Brown 

The nutrition of the Red Algae is much like that of the Phaeo- 
phyceae, except that many of the chemical products of photosynthesis 
and subsequent metabolism are different ; these include the food- 
storage materials, of which a unique starch-like substance is the 
chief. The reproduction is remarkable in lacking any self-propelled, 
flagellate stage. Sexual reproduction is effected by fertilization in 
situ of a large and fixed female cell by a small male one or its dis- 
charged contents — in either case carried along, aimlessly as it were, 
by an impinging water current. Instead of resulting in the formation 
of a new, ' daughter ' plant, fertilization leads to further development 
which results in the formation of special asexual spores called 
carpospores. These in the simpler Red Algae germinate to produce 
sexual plants ; but in the vast majority of types they give rise instead 
to asexual plants producing another kind of asexual spores, called 
tetraspores, which in their turn germinate to produce sexual plants. 
In such cases there is a regular alternation of a sexual generation 
with two asexual ones of which the second is on a separate plant. 

The Red Algae are very widespread. Not only do they occur in 
fair numbers in the habitats occupied by Brown Algae — with which 
they are commonly interspersed even to the extent of frequently 
growing superficially on their bodies, as epiphytes — but there are 
also a number and range of forms inhabiting cool streams and other 
freshwater habitats. Many of the marine types tend to grow in 
deeper water than the Brown or Green Algae, being supposedly 
adapted through their special pigments to photosynthesize far under 
the surface of the water by absorbing the shorter wave-lengths of 
light which penetrate to relatively great depths. Red Algae also 
tend to be more numerous in warm than in cold seas, although not 
a few occur well north in the Arctic. At their best they may dominate 
the deeper layers, especially, of marine coastal vegetation. Of the 
Red Algae, again, the carbohydrates and carbohydrate derivatives 

4 o 




fl.* :/ 

Fig. 9. — Various forms of Red Algae (Rhodophyceae). A, Phyllophora ( X about 

J) ; B, Batrachospermum ( X 5); C, Grinnellia (X about;'); D, Chondrus ( X about 

J); E, Corallopsis ( about |); F, Polysiphonia ( about |). 


are used commercially in the production of colloidal substances that 
are widely employed for food and in industry. Instances are 
' Carrageen ' or ' Irish- moss ', used for food, and agar, which is 
of great importance in bacteriological and allied work, though it is 
even more extensively used in other connections. 

Myxomycetes : These are the Slime-moulds, or Mycetozoa, which, 
as the latter name implies, exhibit animal as well as plant character- 
istics, being indeed near the border-line of the two kingdoms, though 
widely considered as Thallophyta. They are simple organisms 
which in the vegetative condition (i.e., when not reproducing) 
consist of naked, multinucleate masses of protoplasm termed 
' plasmodia '. These show the animal characteristics of slowly 
creeping movement and ingestion of food, and the plant character- 
istics of reproduction by spores (which in some genera have cellulose 
walls) formed in a special spore-producing organ (sporangium). 
The vegetative plasmodium tends to shun the light and to be shape- 
less and often several inches in diameter, growing as long as food 
is available, though when food is scarce it may form a mere starved 
network of living strands, Its outer layer is less liquid than the 
inner portion and is usually devoid of nuclei ; the commonly slimy 
appearance has led to the name of Slime-moulds. Although chloro- 
phyll is lacking, the plasmodium may be variously and often brightly 
coloured — most frequently yellow or brown, but sometimes orange, 
red, black, or even greenish. 

Nutrition of Slime-moulds is primarily saprophytic, the plasmo- 
dium living upon a variety of organic materials such as rotting wood 
and dead leaves, apparently ingesting tiny particles of these and break- 
ing down the complicated carbohydrates in them to simple sugars 
which are used as food. Frequently, living bodies such as Bacteria 
and fungal spores are ingested ; indeed, Slime-moulds can be grown 
experimentally on an exclusive diet of appropriate Bacteria. Fruit- 
ing bodies (sporangia) may be made when food becomes scarce ; 
these are very various in form in different types, as indicated in 
Fig. 10. Often they are gracefully stalked, with rounded or 
elongated sporangia consisting of an outer membrane enclosing a 
mass of very small uninucleate spores and, frequently, a system of 
ramifying tubes. Sometimes almost the whole mass of protoplasm 
becomes a single, large, spore-producing structure. The spores are 
eventually released by rupture of the outer membrane and are 
scattered by the wind. They germinate in water, each spore 

4 2 


producing one to a few flagellated swarin-cells. These, following a 
period of swimming and often of division, become Amoeba-\ikc y and 
after feeding and further division behave as gametes, fusing in pairs 
to form zygotes. The zygotes may grow each into a single Plas- 
modium, or numerous zygotes or plasmodia may fuse together, or 
one plasmodium may divide into two or more. 



Fig. io. — Slime-moulds (Myxomycetes). A, Plasmodium of Didymium ( i); 
B, Sporangia of Hemitrichia (X 15); C, Comatricha (x 20); D, Trichamphora 

(X 10). 

Whereas for spore-formation the plasmodium will usually creep 
to a light and airy place, in general Slime-moulds are found on 
decaying vegetable matter in moist and shady situations. Thus they 
are very widespread in damp woods and thickets, though scarcely 
ever forming any appreciable feature of local vegetation. If the 


group be taken in the wide sense as including also those organisms 
which cause such diseases as club-root of Cabbages and allied plants 
and wasting disease of Eel-grass, it has a considerable economic 
nuisance-value. Otherwise its members, however interesting, can 
hardly be regarded as doing more than a very minor amount of 

Fungi : These are the Mushrooms, Toadstools, Moulds, Rusts, 
Smuts, Yeasts, etc., and comprise, with the Bacteria, the main 
ultimate scavengers of the organic world, besides causing many of 
the worst plant diseases. The Fungi are a large and diverse group 
of relatively simply organized plants. They are usually composed 
of branching tubular filaments (' hyphae ', collectively forming the 
so-called ' mycelium ') and always lack chlorophyll, though occasion- 
ally they may be green in colour. Some are unicellular, and many 
others are microscopic though filamentous ; commonly, however, 
the filaments are sufficiently numerous or massed to be evident to 
the naked eye — usually as a soft whitish investment. They contain 
numerous tiny nuclei and may be divided internally by cross-walls 
(septa), or, alternatively, be non-septate. When reproduction is 
taking place Fungi may be variously, even very brightly, coloured ; 
this is especially the case with some of the larger and often highly 
characteristic fruiting bodies, such as (those of) Toadstools and 
Puff balls, which can reach a considerable size. Fig. 1 1 shows some 
of these fruiting bodies and other reproductive structures of Fungi, 
which exhibit a great diversity of form. In these fruiting bodies the 
masses of filaments, instead of being soft and cobwebby as in the 
Moulds, are so closely interwoven as to form a solid or even hard 
structure of definite organization. 

As they lack photosynthetic pigments and are not, alternatively, 
chemosynthetic, Fungi have to obtain their food for energy and 
body-building by living either parasitically (on or in other living 
organisms) or saprophytically (by the breakdown of dead organic 
materials). As parasites they are the cause of numerous and often 
devastating diseases, especially of plants, and as saprophytes they 
cause widespread decay and effect a large proportion of the breaking 
down of elaborated materials such as leaf-mould. Without such 
breakdown and return of the raw materials into circulation, life on 
earth would be brought ultimately to a virtual standstill, or even 
cease altogether — hence the vital significance of Fungi and Bacteria 
as scavengers. Animal characteristics exhibited by Fungi include 




Fig. ii. — Some Fungi. A, cells of Yeast (Saccharomyces) budding actively 
(X 960); B, Diagram of Mucor mucedo, a common Mould, showing the mycelium 
growing symmetrically from a central spore, and developing sporangia, successive 
stages of which are marked a, b, c, (X about 20); C, Puffballs (Lycoperdon sp.) 
which have opened at the top (x 1%); D, Stereum qffine (x i|); E, a Stinkhorn, 
Ithyphallus (X £); F, Morel (Morchella) (x f); G, Auricularia (x 1); H, the 
Deadly Amanita {Amanita phalloides), a Gill Fungus (X j); I, Boletus, a stalked 

Pore Fungus ( X £). 

the commonly chitinous cell-walls and the storage of food mainly 
as glycogen. Their reproduction may be effected vegetatively by 
separation and subsequent development of part of the mass of 
filaments, and asexually by spores which are usually of very small 
size and produced in enormous numbers (sometimes millions of 
millions by a single fruiting body). In some of the simpler Fungi 
these asexual spores are swimming zoospores ; usually, however, 
they are non-motile and are enclosed in a more or less resistant 
wall. Various sexual processes occur, usually involving the fusion 
of unlike gametes, gametangia, or hyphae, and resulting in the 
production of resting or airborne spores. The ' budding ' practised 
by Yeasts is another effective mode of vegetative propagation (see 
Fig. 11, A, above ; the formation of spores internally is seen below). 
Fungi find habitats for existence almost everywhere there are 
either living organisms to parasitize or dead and decaying organic 
materials to attack. Many are aquatic, including marine, and some 
grow actively even in the absence of free oxygen. The other 
familiar habitats are soils, dung, and various foods, fabrics, and 
wooden or other structures, which Fungi frequently cause to rot. 
Thus they occur throughout the world wherever life is possible 
and they can find materials upon (or in) which to grow. Yet as 


actual components of vegetation their role is usually minor, unless 
it be very locally and temporarily when food supplies are plentiful. 
However, in economic connections their importance is vast and 
multifarious : the frequency with which they cause diseases, especi- 
ally of plants, has already been referred to. Such diseases result in 
many hundreds of millions of dollars' worth of damage to crops 
yearly in North America alone. Also explained above is their 
significance as scavengers, returning the products of organic break- 
down to the air and soil in simple forms that can be absorbed and 
used again by green plants. Many other saprophytic Fungi are 
outstanding nuisances to Man in causing spoilage of food and 
destruction of fabrics, timber, and so forth. Yet others are valuable 
in positive ways, examples being the activities of Yeasts, which are 
employed in the making of wines, beer, and bread, and the use 
of Mushrooms and Toadstools as human food. There are also the 
' industrial ' Fungi that provide valuable sources of certain food 
proteins and vitamins, and the ' medicinal ' ones that provide such 
antibiotics as penicillin. Finally, Fungi play a significant role in 
the nutrition of many higher plants — including forest trees, in or 
upon whose roots they live and form mycorrhizas. 1 

The above nine groups, which we have treated as classes, are 
considered by many authorities to be subdivisions (subphyla) or, 
especially in some instance, full divisions (phyla). The same is true 
of a few other, smaller groups which are of very little importance 
phytogeographically, or as components of vegetation, and which we 
are accordingly ignoring. 

Lichenes : These are the Lichens — peculiar dual organisms 
produced by the intimate association of two plants, a Fungus and 
an Alga or a Schizophyte, and accordingly belonging to different 
groups. They seem best treated as a separate class or subdivision 
of the Thallophyta. Such a ' living together ' for mutual benefit 
is termed a symbiosis, and of this Lichens afford the great example, 
though mycorrhizas (see above) are another. In the formation of 
Lichens the Fungus usually forms a tough, often leathery, invest- 
ment, the algal or schizophyte cells or filaments being interspersed 
or grouped within, most typically forming a layer near the upper 
surface. Lichens are often luxuriant and may live for centuries, the 
symbiosis being evidently a mutually beneficial relationship in that 

1 For an up-to-date account of this intriguing subject, see Dr. J. L. Harley's 
The Biology of Mycorrhiza (Leonard Hill, London, pp. xiv -f 233, 1959). 


the ' algal ' element is enabled, by the protection afforded by the 
fungal envelope, to live in dry and exposed situations where other- 
wise it could not exist, while the Fungus derives food which results 
from the photosynthetic activity of its partner in places where 
otherwise none would be available. 

The ' algal ' elements in Lichens may be members of either the 
Chlorophyceae or the Cyanophyceae and, like the Fungi concerned, 
are usually definitely identifiable. With the varying combinations 
involved, as well as differing heritages, habitats, and growth 
tendencies, a vast array of different forms of Lichens result, though 
their growth tends to be very slow. In size, Lichens vary from 
minute to some which are whole metres in diameter. In colour 
thev may be of almost every conceivable shade, being often of 
brilliant hue ; on the other hand a great many are a dull greenish- 
grey, as a result of combination of the colours of the components. 
The main groups of forms are (1) ' crustose ' (crustaceous), forming 
a thin crust over (or sometimes mainly beneath the surface of) the 
rock or other material on which they grow, (2) ' foliose ', being 
more or less prostrate and flat, with leaf-like lobes, and (3) ' fruticose ', 
being upgrowing, branched, and often bush-like, or pendulous from 
the branches of trees. A range of types is shown in Fig. 12. 

The nutrition of Lichens is primarily by the photosynthesis of 
their ' algal ' components, in which connection the fungal element 
may in a sense be considered parasitic. Vegetative reproduction 
is by fragmentation of the plant body, especially when this becomes 
old and decrepit and liable to break down, or by special structures 
termed soredia, which are groups of fungal filaments interspersed 
with ' algal ' cells. These soredia become separated from the 
parent, being often produced in large numbers, and blow about 
easily. Sexual reproduction is confined to the fungous partner and 
follows its particular pattern, usually involving the formation of 
spores in a special structure, following a fusion of gametes. On 
germination of these spores, new lichen plants are formed only if 
some fragment of the appropriate Alga or Schizophyte is present. 

Lichens occur plentifully in a great variety of habitats, usually 
of dryish nature. Thus they favour the trunks and branches of 
trees, exposed rocks, and bare ground provided the surface is stable. 
In these and other situations they are to be found practically every- 
where on land, though shunning large cities owing to their sensitivity 
to fumes. Although particularly characteristic of high mountain 
peaks and of arctic and antarctic regions, where they may dominate 




or virtually constitute the vegetation over considerable areas of the 
drier habitats, they are also found as minor — but sometimes con- 
spicuous — constituents of the vegetation in temperate and even 
tropical regions. They are especially significant as pioneer colonists 
on bare areas, and in boreal regions where they afford much of the 
winter food of wild Reindeer and Caribou. Economically they are 
important chiefly in the feeding of domesticated Reindeer, on which 

•^.^., 5 #' 


C D 

Fig. 12. — Various forms of Lichens (Lichenes). A, Usnea barbata, a branched 
epiphytic type (x i); B, Haematomma punicewn, a crustaceous type ( i); C, 
Cladonia verticillata, a terrestrial fruticose type (\ I); D, Lobaria pulmonaria, 

a foliose type (X t)- 


whole tribes of northern peoples depend for the wherewithal of life. 
Their plentiful storage of a starch-like carbohydrate also makes some 
of them, such as ' Iceland-moss ', useful for human food, though 
most are highly unpalatable. The use of certain Lichens as sources 
of attractive dyes has greatly diminished with the chemical advances 
of recent years, but some dyes, such as litmus, are still widely 
obtained from them. 


Hepaticae : These, the Liverworts, are a smallish class of usually 
green photosynthetic plants of relatively small size, though always 
multicellular and visible to the unaided eye. There are two main 
forms : the thalloid, having a thin and more or less flat, prostrate 
plant body which tends to branch frequently and equally, and the 
' leafy ', consisting of a creeping central axis up to a few inches 
long, provided with delicate leaf-like expansions. These last are 
only one cell thick and lack a midrib ; they are usually arranged in 
two rows, lying on either side of the often prostrate axis, with 
commonly a third row r of smaller lobes lying along the under surface. 
Noticeable on the lower surface, especially of the thalloid types, 
are numerous thin root-like ' rhizoids ', primarily serving the pur- 
pose of attachment to the ground or other material on which the 
plant grows. Often there are air-chambers or other special features 
on the upper surface. The main photosynthesizing plants are the 
gametophytes, so termed because they produce the gametes ; they 
comprise the gametophytic generation which alternates regularly 
with the spore-producing (sporophytic) one to complete the life-cycle. 

The gametes are formed in minute male and female organs, the 
male spermatozoids swimming freely to fertilize the passive and 
well-protected female ' eggs ', from which, after fertilization, the 
sporophytes develop. In most types these last consist of an absorb- 
ing foot, a more or less elongated stalk, and a roundish capsule 
(sporangium) in which the microscopic spores are produced in 
considerable numbers. The foot is embedded in the tissue of the 
gametophyte, from which it absorbs nourishment. This is passed 
on to the rest of the sporophyte, which in most types lacks chlorophyll 
and is thus parasitic on the gametophyte. In some simple forms 
there is no foot or stalk, the sporangium being embedded in the 
gametophyte, and in one group the sporophyte is photosynthetic 
and grows continuously from near the base. But in any case there 




is a regular alternation of gametophytic and sporophytic generations, 
such as we shall see later in all normal higher plants. Though the 
gametophyte tends to be more evident and ' dominant ', these 
generations in the Liverworts have no independent existence, and 
consequently their dual significance is somewhat difficult to grasp. 
Both generations are depicted in each of the different types of 
Liverworts shown in Fig. 13. 

B C 

Fig. 13. — Types of Liverworts (Hepaticae). A, Marchantia, a thalloid type, 
showing female plant on left and male on right (X il)\ B, C, leafv Liverworts 

(B, x 4; C, x 1). 


The spores, given suitable conditions after liberation, can germinate 
to form fresh gametophytes, so completing the life-cycle. They 
usually afford the main means of multiplication, although vegetative 
methods, such as fragmentation of the gametophyte or formation by 
it of special bud-like bodies called gemmae, may also be effective in 
some cases. Liverworts chiefly inhabit damp places, such as 
stream-banks, sheltered nooks, and the boles of trees and decaying 
fallen branches in shady forests. They also grow on moist soil and 
in tufts of Mosses, etc., and are practically world-wide in distribution 
on land. Many grow on the leaves of other plants in the tropics, 
and some in freshwater habitats. With the exception of a very few 
which are saprophytic, the nutrition of the gametophyte is primarily 
by photosynthesis ; on this the sporophyte is, as we have seen, in 
most instances parasitically dependent. Although they are some- 
times important as pioneers on bare ground and may even form 
more or less ' pure ' patches some yards in extent, Liverworts in 
general play only a minor role as ' fillers ' in higher vegetation. 
Nor have they any particular value for Man, except sometimes as 
aides in the binding and consolidation of eroding surfaces. 

Musci : These are the Mosses, which, in accordance with their 
near relationship, are in many ways closely comparable with Liver- 
worts. The Mosses are all rather small plants in which the gameto- 
phyte, during the greater part (but by no means all) of its life, 
consists of a more or less upright stem bearing small leaves. These 
leaves, unlike their counterparts in leafy Liverworts, usually have 
midribs and are spirally arranged on the stem, which may vary 
from a fraction of an inch to perhaps a foot in length. The midribs 
contain elongated cells, and a central strand in the stem usually 
contains similar elongated cells which are supposed to conduct water 
and nutrients. True roots are absent, but the base of the stem in 
most types is plentifully supplied with anchoring rhizoids. In one 
characteristic and important group, known as Bog-mosses or Peat- 
mosses, the leaf is not only peculiar in lacking a midrib but unique 
in consisting of a network of small living cells separating large dead 
ones which are transparent and perforated, soaking up and holding 
water with extraordinary efficiency — hence the water-retaining 
capacity of many bogs which are largely formed by such plants. 

On the gametophyte are borne the minute male and female organs, 
commonly in groups that are made evident by the modification of 
the surrounding leaves, and either on the same (hermaphrodite) 




plant or, more often, on separate (male and female) individuals. 
Fertilization is again by a motile spermatozoid which, when water 
is present, swims to the passive and protected egg. The body 
formed by this sexual fusion develops into the sporophyte which, 
when mature, consists of an absorptive foot, a usually long stalk, 
and a more or less complicated and characteristic capsule. Fig. 14 
shows both the gametophyte and sporophyte of different types of 
Mosses, for here again the two generations are unseparated, forming 
one continuous plant body, the sporophyte being at first parasitic 
on the gametophyte but later becoming at least partly self-supporting 
through its possession of chlorophyll. 

The spores, formed in the capsule and liberated often by some 
complicated mechanism that works only in dry air (which is more 
beneficial than moist air for their further dispersal), are again the 
main mode of multiplication. But instead of growing directly into 
a typical new gametophyte as in the Liverworts, they develop in the 
Mosses, on germination, into an extra stage known as a ' protonema'. 
This is an independent, cellular plant containing chlorophyll and 
manufacturing its own food. It is filamentous and branched in 
most types but in some it is thalloid, like the gametophyte of many 
Liverworts. On the protonema develop lateral buds which grow 




Fig. 14. — Some Mosses (Musci). A, gametophyte of a Moss, showing a group 
of male and female organs at the top ( x 7) ; B, a typical species of Sphagnum, 
the Bog-mosses (x about 4); C, Atrichum (Catharinea), showing at the base the 
masses of rhizoids and protonemal filaments from which grow the leafy gameto- 
phtyes bearing, above, the sporophytes ( X 2). 

into the main, leafy gametophytes ; thus is the life-cycle completed. 
Mosses can also propagate by gemmae and multiply by fragmenta- 
tion, and they are remarkable for their power to remain alive after 
long periods of desiccation. Their nutrition is again primarily by 
photosynthesis, starch being stored, although some appear to be 
partially saprophytic. 


Mosses grow on a wide variety of exposed surfaces — particularly, 
but by no means entirely, in damp situations. Thus they occur 
plentifully on the ground, on tree-trunks in moist woodlands, on 
decaying wood, on old brick and stone structures, on rocks and 
boulders, and in both still and running fresh water. They are also 
common as subsidiary forms in higher vegetation. Mosses are 
more numerous in species and individuals than Liverworts, and 
tend to cover considerably larger areas and to be far more conspicuous 
— especially in arctic and boreal regions, and high up on mountains. 
They are relatively important as components of natural vegetation 
and frequently dominate substantial areas especially of bogs, whose 
water-level they often raise. In so doing they may even destroy 
tracts of forest and make terrain difficult to traverse, thus affecting 
the economy of Man. On the positive side they are important as 
producers of peat, which often consists largely of the remains of 
Bog-mosses, and as stabilizers of sand-dunes and other erosive 
systems whose surfaces they help to bind. Peat is used extensively 
as fuel and in the improvement of soils. Owing to their insulating 
properties when dry, Bog-mosses are also used in construction work 
and packaging, and, owing to their absorptive and water-retaining 
powers, for surgical dressings and the transport of living plants. 

The above groups all belong to the non-vascular cryptogams and 
lead up to the vascular plants {Vascular es, or Tracheophytes), to 
which all the remaining groups belong. The vascular plants are 
those possessing a vascular system, and include all the most advanced, 
or evolutionary ' higher ', types ; these are generally the largest 
and most dominant on land. A vascular system consists mainly of 
special tracts (bundles) of elongated wood (xylem) and ' bast ' 
(phloem) cells forming a continuous system linking all the main 
parts of the plant. Its chief manifestations are such bundles in the 
stem and veins in the leaf. The primary purpose of such a vascular 
system is the conduction of water, mineral salts, and elaborated food 
materials to portions of the plant where they are needed, so that it 
is partly comparable with the blood and lymphatic systems of higher 
animals. Its secondary function is to give mechanical support, 
especially in the ' secondarily thickened ' older stems of perennial 
plants which consist largely of vascular tissues and have the whole 
crown to support. An account of the general make-up of a vascular 
plant was given in Chapter I, with some indication of how ' The 
Ideal Plant ' lives and grows (pp. 12-14). 


the various groups of plants 


Equisetineae : These are the Horsetails, of which the living 
examples are mere depauperated relics of a group which was much 
more important in earlier geological ages, when it included larger 
tree-like forms. Those remaining belong to a single genus of 
perennial, herbaceous plants consisting of an underground stem 
(rhizome) beset with fibrous roots, and sending up usually erect and 


-Field Horsetail (Equisetum arvense agg.). A, sporophyte with fertile 

branch (on right) and two young sterile branches ( X § ) ; B, mature sterile branch 


stiff, grooved aerial stems that are generally slender but hollow 
and bear at the nodes (parts of the stem where leaves arise) close 
whorls of rudimentary scale-leaves. The stem is commonly green 
and photosynthetic, rarely more than a few feet high, and either 
unbranched or, more often, bears whorls of slender branches (which 
may themselves be much-branched) in the axils of the scale-leaves. 
Fig. 15 shows a characteristic modern Horsetail. Such plants 


practise the mode of nutrition of normal vascular types, namely, 
photosynthesis together with absorption of the necessary elements 
(usually in simple compounds) from the air and soil, or, when the 
plants are aquatic, in solution with the water in which they live. 
In common with almost all of the plants remaining to be described, 
the main food-storage substance is starch. 

These relatively large plants are the sporophytes, which comprise 
the main, dominant generation in this and all the remaining groups. 
The spores are produced in special organs (sporangia) developed on 
short outgrowths which are compacted into distinct, cone-like 
1 strobili ' that are commonly developed at the tops of the stems. 
The spores are all alike but unique in having attached to them at 
one point four slender processes that bend or straighten rapidly with 
changes in atmospheric humidity, consequently often causing the 
spores to move. On germination the spores produce a gametophytic 
body called a prothallus. This is small and possessed of rhizoids, 
irregularly branched, green and photosynthetic, and usually each 
one produces organs of only a single sex. After fertilization by the 
peculiar spiral, multiflagellate swimming spermatozoid, the egg 
develops in situ into a young sporophyte plant which soon becomes 
independent, so completing the life-cycle. 

In spite of the limited number and variational range of living 
forms, Horsetails occupy a considerable array of land and freshwater 
habitats and are geographically very widespread, extending from the 
tropics to the highest latitudes of land. They especially favour 
marshes, lakesides, and damp sand or silt, which they may colonize 
aggressively ; but as components of more mature vegetation they 
are very minor. The outer part of the stem is often heavily 
impregnated with silica, which has led to some species being widely 
used for scouring pots and pans — hence their alternative name of 

Lycopodineae : These include the living Club-mosses (Ground- 
pines), Spike-mosses, and Quillworts, as well as numerous huge 
trees of earlier geological ages that are now known only as fossils. 
The present-day representatives are lowly and herbaceous, differ- 
entiated into stem, roots, and leaves, the last being numerous and 
usually small as well as simple. The stems are rarely more than a 
matter of inches in height or feet in length. Most species are 
evergreen, overwintering without the aerial parts dying back. The 
stems are trailing or upright and usually branched, or, in the Quill- 


worts, unbranched and extremely short ; these last, peculiar types, 
also have relatively large, elongated leaves. In them and some other 
members, the leaf has a characteristic tongue-like appendage (ligule) 
near its base. Fig. 16, showing examples of some types from this 
group, indicates the range of existing forms. These are the main, 
sporophytic plants, and they produce spores in sporangia borne 
singly in the axils of special and usually modified leaves (sporophylls) 
that either occur in groups at intervals along the stem or, more 
often, form terminal cones. In some types the spores are all of 
one kind, but, in others, different sporangia produce numerous 
small ' microspores ' or relatively few (occasionally one) large 
1 megaspores '. 

On germination the spores produce prothalli, representing the 
gametophvte generation. These are always small and relatively 
obscure bodies. However, they vary in different types from lobed 
photosynthetic ones or underground non-green tuberous ones living 
saprophytically with the aid of mycorrhizas (in either case usually 
producing both male and female organs on the same prothallus), to 
limited growths largely enclosed within the old spore-wall in those 
instances where spores of two different sizes occur. In such instances 
the megaspores each produce a few female organs and the tiny 
microspores only a single male organ, the ' prothalli ' being dependent 
upon the food stored earlier in the spore by the sporophyte. Follow- 
ing fertilization by the spirally-shaped, swimming spermatozoid, the 
egg develops in situ into a new sporophyte plant which is photo- 
synthetic and independent from an early stage, so reversing the 
situation met in Bryophyta. Some Club-mosses produce bulbils or 
gemmae which are effective in multiplying the sporophyte. More- 
over, vegetative propagation following fragmentation of large old 
plants often occurs, at least among the longer trailing types. 

The Lycopodineae are fairly numerous in species and wide in their 
habitat tolerance. Though most characteristic of shady woods from 
tropical to boreal regions, or, in the case of Quillworts, of the beds 
of freshwater lakes, they also occur in more exposed heaths and 
marshes and, in such situations, range far north in the Arctic. 
Nevertheless, as components of vegetation, except very locally and 
then usually far beneath the dominants, or occasionally in deserts 
where little else grows, they are so very minor as to be almost 
negligible. Very different was the position of some of their fossil 
relatives, which, as we shall see in Chapter V, apparently dominated 
whole forests in much earlier geological ages, and greatly contributed 

3 ), all Club-mosses (Lycopodium species); D (x i), a Spike-moss (Sclaginello); 
E ( X about I), a Quillwort (Isoetes). 



to the formation of coal. Otherwise their economic significance is 
very limited : some warmth-loving species are used as pot-plants, 
and trailing Club-mosses are made into Christmas-wreaths (hence 
another name, ' Christmas-greens '), while the minute spores of 
members of the same genus are so highly inflammable owing to 
stored oil that they can be used to produce ' stage lightning \ They 
are still employed in dusting operations where a fine powder is 
required, and for demonstrating sound-waves in physics. 

Filicineae : These are the Ferns, and include the majority of 
living Pteridophyta as well as some extinct forms. They are 
perennial plants, sometimes small and moss-like but usually of at 
least substantial size. The stems range from creeping and slender 
to erect and stout, and from subterranean to aerial or occasionally 
aquatic ; in Tree-ferns, the often massive, erect aerial stems may 
be several yards high. There are usually plentiful fibrous roots 
below, and, above, leaves (fronds) that characteristically are large 
and compound, composed of more or less numerous segments. 
Occasionally, however, the leaves are small and simple ; indeed the 
Ferns are very varied in form, as may be seen from Fig. 17, which 
shows a range of different types. These, of course, are all sporo- 
phytes, the gametophytes being always small and insignificant. 

The spores are formed in sporangia which are commonly borne 
in groups upon or partly within the lower surface of the ordinary 
leaves, though in many cases the spore-producing leaves, or parts 
of leaves, are modified — sometimes so drastically that they are 
scarcely recognizable as leaf members. In most types the spores are 
all of one kind and produce on germination a filamentous, or more 
often a flat, green prothallus rather like a small unbranched thalloid 
Liverwort, anchored to the ground by rhizoids and bearing both 
male and female organs, though a few types have a subterranean 
and saprophytic prothallus. However, in the small and usually 
aquatic group known as Water-ferns, which have slender stems and 
small and sometimes very simple leaves, separate microspores and 
megaspores are formed, which produce male and female organs, 
respectively, on germination. Following fertilization by the swim- 
ming, corkscrew-like spermatozoid, the egg develops into a fresh 
sporophyte plant which soon becomes photosynthetic and inde- 
pendent. This is its primary mode of nutrition. Thus the life- 
cycle, involving as usual in these vascular land-plants an alternation 
of sexual and asexual generations, is completed. The sporophytes 





Fig. 17. — Various types of Ferns (Filicineae). A, Shield-fern (Dryopteris) 
( 5); B, a group of Tree-ferns, Cyathea (scale indicated by man in foreground); 
C, Pteris longifolia ( X rfr) ; D, Moonwort (Botrychium) ( X about |) ; E, Common 
Adder's-tongue (Ophioglossum vulgatum) (X about |); F, a Water-fern (Marsilea) 

( X about ^). 

of many Ferns also propagate vegetatively — for example, when the 
old parts of types with branching rootstocks die off and fragmentation 
results, or through the growth, after detachment, of special bulbils 
or plantlets that develop on the leaves of some species. 


Although mainly favouring shady and humid situations, Ferns 
occupy a wide array of habitats ranging from dryish heaths and 
crevices of rocks to wet mud and open fresh water, and from forest 
floors to quite lofty branches or crutches of trees. They are plentiful 
especially in the damper tropical and temperate regions and reach 
the southern portion of the Arctic in fair array, while a very few 
species persist northwards to almost the highest latitudes of land. 
Vegetationally they are chiefly of importance as subsidiaries in humid 
habitats from the tropics northwards and southwards to temperate 
regions, though arborescent types may contribute substantially to 
the forest, for example in New Zealand. North of the temperate 
regions in the northern hemisphere they tend to become scarcer, 
being often absent from dry areas and almost negligible as com- 
ponents of vegetation in most boreal and arctic regions. Their 
economic significance is chiefly aesthetic and horticultural ; many 
are among the most beautiful of living things, and consequently 
fern-growing is a popular hobby. Some are minor constituents of 
animal fodder or may be employed as food by humans or cut and 
dried for litter, but probably far outweighing these uses is the 
nuisance- value of others — particularly the common Bracken, which 
is a pestilential weed that widely overgrows pastures and young 
tree plantations. 


Gymnospermae : This, the more primitive of the two classes (by 
some considered subdivisions) of the Spermatophyta (Seed-plants), 
includes the Conifers, Cycads, and Gnetales among living groups, 
and many extinct fossil representatives that were of great importance 
as components of vegetation in earlier geological ages (see Chapter V). 
The Seed-plants are the main phylum existing on land today, 
providing the vast majority of dominant species and the great 
preponderance of vegetation in most situations. They are, of 
course, vascular plants, being, briefly speaking, those which bear 
seeds. A seed is an organ peculiar to this ' highest ' division of the 
plant kingdom and is the product of a fertilized ovule, consisting 
of an embryo that is often embedded in a nutritive tissue and is 
normally enclosed by one or two protective seed-coats. The large, 
complicated visible plant is always the sporophyte generation, the 
usually microscopic female gametophyte being embedded within it 
and never having a separate existence, while the male gametophyte 
is even more reduced. 


The Gymnosperms are distinguished by having ovules which are 
borne ' naked ' on leaf-like organs (cone-scales). Although not 
enclosed in an ovary, the ovules are often well protected by the 
mutual contact of the cone-scales or by the development of other 
special structures. For fertilization a small gap is left or an opening 
occurs, and for liberation of the ripe seeds the cone-scales or other 
protective structures simply spread apart. The living Gymnosperms 
are all perennial woody plants, ranging from small shrubs to the 
very largest of trees. Their internal structure is generally different 
from, and more primitive than, that of the other, remaining group ; 
and they are far less numerous in species and individuals, tending 
to be less widely dominant. They show, however, effective internal 
differentiation and external division of labour, allowing conduction 
to the tops of the world's tallest trees (Coastal Redwoods, which 
reach 364 feet in height 1 ), and, externally, involving highly specialized 
roots, stems, leaves, and intricate organs of reproduction. The 
roots are generally much-branched and fibrous, the stems essentially 
columnar at least below (though they may vary from tuberous in 
some Cycads to much-branched and shrubby, especially in Ephedras), 
while the leaves vary from very large and subdivided in Cycads to 
small and needle-like in many Conifers. The plants most commonly 
grow as trees, the leaves being usually evergreen, lasting for several 
years. Examples of Gymnosperms are shown on pages 64-67. 

The structures that ultimately produce the male and female 
gametes are borne on modified leaves, usually aggregated into terminal 
cones of one sex and developed either on the same (hermaphrodite) 
or on different (male and female) plants. The female gametophyte 
consists of a mass of cells developing within the megaspore and, 
like it, remaining hidden in the sporophyte. In this gametophyte 
develops the egg, which, with the immediately surrounding envelopes 
comprising the ovule, forms the seed after fertilization. Sexual 
fusion is effected by male gametes which may be motile spermato- 
zoids or merely passive nuclei, and which are normally enclosed 
within a pollen tube growing towards the egg ; these gametes are 
formed by a microscopic and vestigial male prothallus developed on 
germination of the microspore. This last is the pollen grain, 
produced in great numbers in special organs, the pollen sacs, and 
each with some infinitesimal chance of being carried by the wind 
to the vicinity of a receptive ovule. The seed contains the embryo 
sporophyte and, after liberation from the parent and given suitable 

1 This goes for living trees, though some Eucalypts in south-eastern Australia 
appear to have exceeded this figure in the recent past — see note on p. 379. 

6 4 


conditions, grows into the new generation, so completing the life- 
cycle. Thus, in such sexual reproduction-cycles, the chances of 
dispersal of individuals are limited to the seed, although the pollen 
grains can transport heritable characters. Nor do Gymnosperms 
normally possess effective modes of asexual reproduction, except for 
such vegetative methods as ' layering ', which involves the rooting 


of lateral shoots or branches and their separate growth on segregation 
from, or death of, the parent. 

Among the Gymnosperms, the usually stocky and unbranched 
Cycads, with their palm-like crown of huge compound leaves, are 
chiefly characteristic of the drier areas of the tropics and subtropics, 
as are the much-branched, bushy Ephedras, though their broad- 
leafed relatives, the Gnetums, favour moist tropical habitats. Far 
more numerous, important, and widespread, however, are the 
Conifers, different members of which occupy almost the complete 
range of land habitats from swamps to dry sands. They dominate 
vast areas of temperate and boreal forests, many of which they 
virtually compose, and constitute the northern limit of arborescent 
growth practically around the top of the globe, as well as, often, 


{See p. 67.) 


the altitudinal limit on mountains. Altogether they probably play 
a role in constituting higher vegetation on land, which is second only 
to that of the other, last remaining group which we shall discuss 
next. Moreover, their economic importance is in keeping with such 
a position, for besides affording shelter and greatly affecting Man's 
environment, they provide him with a large proportion of his timber, 
pulpwood, turpentine, firewood, and numerous other commodities, 
besides minor foods and other items of everyday or local life too 
numerous to mention. 

(See p. 67.) 



6 7 

Fig. 18. — Some examples of Gymnosperms (Gymnospermae). A, a Cycad 
{Cycas rumphii) ( X -it) ', B, male twig of Ephedra ( X f ), a member of the Gnetales ; 
C, twig of typical Conifer (Pinus insularis), showing female cones of the three 
latest years (x £); D, Coastal Redwood {Sequoia sempervirens), another Conifer 
(scale indicated by standing man). (Phot. W. S. Cooper.) 

Angiospermae : These, the flowering plants, are evolutionarily 
the highest, and vegetationally and economically the most important, 
of all groups in the world today. They are seed-plants, but dis- 
tinguished from Gymnosperms by having their seeds enclosed in an 
ovary — a variously shaped, but commonly roundish, vessel formed by 
the enclosing ' carpel ' (or ' fused ' group of two or more carpels) 


which produces the seed or seeds internally. After fertilization, the 
ovary becomes the fruit. The Angiosperms are also generally to 
be distinguished from the Gymnosperms by their internal structure 
and by their possession of flowers, which are specialized short 
reproductive shoots bearing typically four different sets of organs 
in close proximity. These are (i) on the outside the sepals, which 
are usually leaf-like and protective in the bud stage, and inside of 
which come (2) the petals, which are commonly attractive in colour, 
form, and odour ; then (3) the stamens, producing the pollen grains 
(microspores), and finally (4) the one or more carpels lying in the 
centre and producing the ovule or ovules. 

Various types of Angiosperms are so entirely familiar to us all 
that it would be superfluous to illustrate them here. Instead, Fig. 19 
(pp. 70-71) shows a diagrammatic representation of a dicotyledon- 
ous (see p. 73) Angiosperm flower, and, in addition, sections of 
stems of monocotyledonous (see p. 73) and dicotyledonous plants 
to indicate the disposition and something of the appearance of the 
vascular bundles when magnified. Many examples of Angiosperms 
will be found illustrated in the chapters on vegetation (especially 
Chapters XII-XIV), and of their fruits and seeds there are accounts 
in Chapter IV, whilst the two main groups of them, Monocotyledons 
and Dicotyledons, are distinguished in the last paragraph of the 
present chapter. Familiar Angiosperms include all of our common 
agricultural and garden crops, all Grasses and other flowering herbs 
whether annual or perennial, and almost all broad-leafed trees and 
shrubs such as Oaks, Elms, Beeches, Maples, Birches, Poplars, and 
Willows. There is consequently no need to emphasize that each 
consists primarily of roots, stem or stems, and leaves, nor to describe 
the form of these organs. 

Sexual reproduction is the object of the flowers and with it the 
stamens and carpels are particularly concerned. Frequently the 
stamens and carpels are in difTerent flowers or even on different 
plants, in which event the individuals are unisexual. In any case 
the function of the stamens is to produce the pollen grains, usually 
in large numbers, for transference (by such agencies as wind, animals, 
or water) to the stigma, which is the receptive part of the carpel. 
Within this last the ovules are formed, each containing a female 
gamete. The pollen grains germinate on the stigma, sending out 
pollen tubes which come to contain the male gametes and usually 
grow through the underlying tissue, deriving nourishment as they 
go. This growth of pollen tubes normally goes on until the tip of 


one reaches the immediate proximity of the female gamete in the 
ovule, there discharging the male gametes, one of which effects sexual 
fusion. From the fertilized ovule the seed develops, contained in 
the ovary which commonly becomes the fruit. Seeds and fruits, 
though often alike in appearance, are technically very different and 
should always be distinguished. Many fruits are attractive to 
animals, or are winged or plumed to be caught in the wind, or 
buovant to float on water, being dispersed by these agents with the 
seed inside. However, in those fruits which contain more than one 
seed, it is more effective to liberate the seeds and have these individu- 
ally attractive or appendaged for separate transportation. In any 
case the embryo within, if still alive and given the right conditions, 
germinates to form a new sporophyte plant which soon becomes 
independent of any stored food-reserve and so completes the life- 
cycle. Such is the general story — though there are all manner of 
variations and even exceptions — in which it should be noted that 
the gametophytes, both male and female, are microscopic, vestigial, 
and entirely dependent on the sporophyte for food, the female being 
so embedded therein that to all appearances there is only the one, 
sporophyte generation. 

Asexual reproduction is extremely common and widespread in 
Angiosperms. Not only are there numerous species which are 
habitually parthenogenetic, the ovules developing successfully with- 
out fertilization, but there is a wide array of vegetative means of 
propagation in nature quite apart from those commonly practised 
by Man. Familiar examples are the underground stems (rhizomes 
and rootstocks) as well as suckers and overground runners and stolons 
of many plants which, rooting at the nodes, constitute daughter 
individuals on severance from or death of the parent. Also familiar 
is the fragmentation of many water or colonial plants, as well as 
separation of bulbs and tubers, while the production of bulbils or 
young plantlets in place of flowers in many species, or in the axils 
or even on the margins of leaves in others, affords further ready 
means of vegetative propagation. Indeed, so common and effective 
are these or other asexual methods, that many plants resort to them 
habitually, and frequently are enabled by employing them to 
reproduce and live indefinitely in regions where climatic or other 
conditions prevent the ripening of fruit or even successful flowering. 

The mode of nutrition of most flowering plants is primarily by 
their own photosynthetic activity, which takes place mainly in their 
green leaves. Here, with the aid of chlorophyll in the light, they 



Pith ray 



Vascular bundle 

Pith ray 


Endodermis «.- 

: ?> Sclerenchyma 

Fig. 19. — Features of Angiosperms. A, diagram of section of flower at time of 
fertilization, showing pollen grains (greatly magnified) germinating on the stigma 
and pollen tubes growing down towards ovules (seen in centre). Ranged around 
are the stamens, the large attractive petals, and the sepals which protected them 
when in bud. B, cross section of stem of a monocotyledonous Angiosperm {Zea 
mays) ( X about 30) ; the dark oval areas are sections of vascular bundles which 
are characteristically scattered (instead of being disposed in a more or less peri- 
pheral ring, as in most Dicotyledons). C, diagram of portion of stem of a dicoty- 
ledonous Angiosperm dissected in cross, radial, and tangential sections to show 
the various tissues of which it is composed ( X about 80, but not all parts to precisely 

same scale). 



build up simple raw materials to complex carbohydrates. But 
besides carbon dioxide and water they also require, like the lower 
plants, certain mineral elements which are usually obtained from the 
soil or water in which they grow ; alternatively, some other materials 
may be objectionable or even poisonous to them. The likes and 
dislikes, as well as requirements and inabilities, of different plants 
in the matter of water or soil constituents, often complicate their 
distribution patterns, and constitute, as we shall see later, one of 
the many sets of factors determining their actual or potential areas 
on earth. Not a few of the Angiosperms which are primarily 
photosynthetic appear to be aided in their nutrition by mycorrhizal 
associations with Fungi in their roots — examples being the Heaths, 
Orchids, and many forest trees. It has even been suggested that 
the majority of vascular plants may be so aided. In addition there 
are entire groups of Angiosperms that are either wholly or partially 
parasitic on other plants, or saprophytic on a variety of decaying 
substrata — again largely with the aid of mycorrhizas. 

As for their habitats, Angiosperms tend to be plentiful almost 
everywhere anything can grow on land or in shallow fresh water, 
though they may be scarce in, or even absent from, some of the most 
inhospitable situations such as rock faces, deserts, or mountain 
summits where nevertheless some lower plants may exist. Except 
in warm regions where free-floating types flourish, they do not 
normally occur in deep open water ; nor do they grow directly on 
snow or ice, as some lower organisms can. Relatively few live in 
the sea, and these appear to be limited to rather shallow water and 
in no instance to extend northwards beyond the low-arctic zone. 
But, in general, Angiosperms are practically ubiquitous in anything 
approaching orthodox situations for plant growth, and a recital of 
their habitats would be practically that of plants in general as outlined 
in Chapter XI. Within the limits stated they are also virtually 
cosmopolitan, extending from the tropics to the farthest north land, 
which many attain, and also to the Antarctic Continent, which does 
not appear to be reached by any other vascular plants nowadays. 
Moreover, it seems likely that in the matter of number of species 
they may be the largest of all plant groups, the order of 250,000 
being currently suggested, though of course a great deal depends 
on what precisely is understood by a species. 

Finally, in the matter of vegetational and economic significance, 
Angiosperms are of paramount importance in the world today, 
affording the main dominants of most plant communities on land 


and of many in the water, and comprising almost all agricultural 
and horticultural crops as well as the majority of forestral products. 
To one or other of these aspects the remainder of this book will 
bear such abundant testimony that it would be superfluous to give 
details here, though the reader may be referred especially to Chapters 
XII-XVT for vegetational aspects and to Chapter IX for economic 
ones. It is by the distribution and growth potentialities of angio- 
spermic plants, more than any other group, that human migrations 
have been affected in the past, and civilizations have been caused 
to wax and wane. 

In view of their general importance it seems desirable here to 
point out the two main groups (subclasses if the Angiosperms be 
considered a class) into which the latter are usually divided (though 
the individual criteria are not infallible). These are the Mono- 
cotyledones (Monocotyledons), characterized by having a single 
seed-leaf (cotvledon), and the more numerous Dicotyledones 
(Dicotvledons), characterized by having two seed-leaves. In addi- 
tion, the Monocotyledons usually have (a) narrow leaves with 
parallel veins, and (b) the vascular bundles in the stem loosely 
scattered and unable to extend ; also (c) rarely any woody develop- 
ment, and (d) the flower-parts most often in whorls of three. The 
Dicotyledons, on the other hand, usually have (a) broad foliage 
leaves with net-like veins, and (b) the vascular bundles disposed in a 
ring in the stem and commonly able to extend indefinitely ; also 

(c) often extensive woody development to form shrubs or trees, and 

(d) the flower-parts most commonly in fours or fives. Examples 
of Monocotyledons are the Grasses, Sedges, Aroids, Orchids, Palms, 
and Lilies ; of the Dicotyledons, most forest trees (other than 
Conifers, Palms, etc.), members of the Pea family, and such crops 
as Beets, Cabbages, Tomatoes, and Cucumbers, in addition to the 
majority of broad-leafed herbs and shrubs. 

Further Consideration 

Many more details and illustrations of each of the main systematic 
groups of plants may be found in almost any modern textbook of general 
botany, such as R. D. Gibbs's Botany ; an Evolutionary Approach (Blakis- 
ton, Philadelphia & Toronto, pp. xiii -f- 554, 1950), or R. C. McLean & 
W. R. Ivimey-Cook's Textbook of Theoretical Botany, vol. I (Longmans, 
London etc., pp. xv -f 1069, 195 1) — or, for the predominant Angiosperms, 
vol. II (ibid., pp. xiii -j- 1071-2201, 1956). It should, however, be 


remembered that authors rarely agree as to, the status and disposition, 
or even the limits, of every group. 

An attempt to cover all the groups is made in A. Engler & K. Prantl's 
Die natiirlichen Pflanzenfamilien, second edition (Engelmann, Leipzig, 
or latterly Duncker & Humblot, Berlin, numerous volumes from 1924), 
and, in greater detail, in A. Engler's very incomplete Das Pflanzenreich 
(formerly published by Engelmann, Leipzig). Also primarily under 
Engler's name are published from time to time revised editions of the 
handy Syllabus der Pflanzenfamilien (Borntraeger, Berlin), giving an out- 
line of the entire plant kingdom. 

For further details the following treatments of the various groups 
should be consulted : 

C. E. Clifton. Introduction to the Bacteria, second edition (McGraw- 
Hill, New York etc., pp. xiv + 558, 1958). 

K. V. Thimann. The Life of Bacteria : their Growth, Metabolism, and 
Relationships (Macmillan, New York, pp. xviii -f- 775, 1955). 

F. E. Fritsch. The Structure and Reproduction of the Algae (Cambridge 

University Press, Cambridge, Eng., vol. I, pp. xvii + 791, 1935 
and vol. II, pp. xiv -f- 939 and 2 additional maps, 1945). 

G. M. Smith (ed.). Manual of Phycology (Chronica Botanica, Waltham 

Mass., pp. xii + 375, 195 1) ; also Algae. 

E. A. Gaumann & F. L. Wynd. The Fungi (Hafner, New York & 

London, pp. 1-420, 1952). 

C. J. Alexopoulos. Introductory Mycology (Wiley, New York, pp, 

xiii -f 482, 1952) ; also Fungi. 
A. L. Smith. Lichens (Cambridge University Press, Cambridge, Eng. 

pp. xxviii -j- 464, 1921). 

F. Verdoorn (ed.). Manual of Bryology (Nijhoff, The Hague, pp 

ix -f 486, 1932) ; Bryophytes. 

F. Verdoorn (ed.). Manual of Pteridology (Nijhoff, The Hague, pp 

xx -f 640, 1938) ; Pteridophytes. 

G. M. Smith. Cryptogamic Botany, vol. II, Bryophytes and Pteridophytes 

second edition (McGraw-Hill, London etc., pp. vii -J- 399, 1955). 

C. J. Chamberlain. Gymnosperms : Structure and Evolution (University 
of Chicago Press, Chicago, 111., pp. xi + 484, 1935). 

G. H. M. Lawrence. Taxonomy of Vascular Plants (Macmillan, New 
York, pp. xiii + 823, 1951); Angiosperms, etc. 

Chapter III 


Plants can grow only where the conditions are reasonably suitable 
for them, and different species have often entirely different needs. 
It follows that local conditions are a primary factor in limiting the 
distribution of any particular kind of plant. Tropical plants cannot 
survive in arctic conditions, nor aquatic plants in a desert. The 
same holds good to varying degrees in less obvious instances, down 
to examples where the balance is so fine that the difference between 
success and failure, or actual life and death, is struck by some 
barely perceptible difference in local conditions. Very often a com- 
plex of interacting factors will be found operating, whose differences 
may be extremely small but nevertheless sufficient to determine 
whether or not a particular plant can grow successfully in a given 

What actually determines the reactions of a plant to the conditions 
making up the environment in which it finds itself ? Fundamentally 
it is the general physiological make-up of the kind of plant involved, 
although the state of development of the individual and its degree 
of adaptation to local conditions may also come into play. Plant 
physiology deals primarily with the internal workings of plants, 
whether biological, chemical, or physical. The main physiological 
characteristics that are found in a particular species are usually 
inherited and, taken together, largely determine the conditions under 
which it can grow and the places where it can survive. 

Physiological Make-up 

Later in this chapter we shall give an account of special features 
which enable plants to offset or at least limit the effects of unfavour- 
able conditions. Such so-called ' adaptations ' include physiological 
acclimatization, and are commonly responses to external conditions. 
When not inherited they seem best considered as mere temporary 
modifications in make-up. Their real value to the plant can be 



seen from such obvious examples as the tall stems of forest plants 
which reach for light, or the long roots of desert plants which seek 
out water. 

Before considering these adaptive modifications, we must deal 
with those physiological attributes which are most significant in 
affecting plant distribution, for they are fundamental to plant 

Water is essential for the life and growth of plants, being a con- 
stituent of their bodies and necessary for many of their life-processes. 
Consequently its availability is among the most important factors 
of a plant's environment. Where there is no water, plants cannot 
long persist in an active state ; although seeds may survive bone- 
dry for many years, they need water to germinate as the plants do 
to grow. Between a desertic lack and an aquatic superabundance 
of water, which are extremes that only suitably adapted plants can 
withstand, there are various degrees of water availability to which 
particular species are accustomed and often limited. The need for 
water largely determines the distribution of plants on the face of 
the earth, as we may see when passing from any lastingly dry area 
to a wet one, when the flora and vegetation will change drastically. 

The actual effect of available water may be complicated by con- 
ditions, such as temperature and atmospheric humidity, that affect 
its utilization within the plant — for example through controlling 
absorption by the roots, movement in the stem, or loss from the 
leaves, etc. Particularly susceptible to atmospheric changes are the 
microscopic pores (stomata) through which most water-vapour and 
other gaseous exchange takes place between the internal tissues of 
higher plants and the atmosphere at large. Consequently the all- 
important water economy of the plant is affected by conditions in 
the surrounding air as well as by the availability of water in the 

Temperature is another of the most important factors of the 
plant's environment. Particular plants require particular tempera- 
ture-ranges for their life-processes and normal development, and, 
different temperatures being characteristic of different climates, such 
requirements widely limit the geographical distribution of plants. 
And along with plants, of course, go the vegetation-types which 
they make up. 

Under otherwise constant conditions each plant has an optimum 
temperature at which it does best, and, on either side of this, a 
range extending to maximum and minimum temperatures beyond 


which it cannot grow normally and many even be killed. However, 
for most plants, in the words of one specialist correspondent, 
1 temperature requirements depend on illumination ; in low light 
[intensities] the optimum is cooler than in high light \ In nature, 
temperatures fluctuate more or less markedly and affect different 
life-processes differently, so that the optimum must take into 
consideration such natural fluctuations on one hand and the optima 
for different life-processes on the other. Even the maxima and 
minima, outside of w r hich death may result, often vary with other 
phvsical factors and with the recent experience as well as evolutionary 
historv of the plant in question. They may also vary with the 
time of exposure as well as with the state of the plant structure or 
its stage of development. Thus resting seeds and other reproductive 
bodies are, in general, far more resistant to extremes than are adult 
plants or, particularly, tender young parts : whereas the killing of 
young shoots and blossoms by even the slightest frosts is an all-too- 
common experience in temperate regions, more mature parts of the 
selfsame plants often survive. Indeed there are numerous known 
instances, involving all the main groups of plants, of such resistant 
bodies as spores and seeds surviving much lower temperatures in 
laboratories than are ever found in nature — including those of liquid 
hydrogen or even of liquid helium near absolute zero. 

Far from all vital activity ceasing at the freezing point of water 
(32 F. = o° C), there are known instances of such physiological 
functions as photosynthesis and respiration proceeding at tempera- 
tures below this point in higher plants, w r hile some Bacteria and 
Fungi are capable of growth at temperatures as low as i6° F. 
(— 8-89° C). It has even been claimed, in Russia, that flagellate 
Algae have been observed swimming in drops of brine cooled 
artificially to — 15 C. On the other hand, whereas most plant 
bodies are killed by heat at much lower temperatures than the boiling 
point of water at sea level (ioo° C), some bacterial spores are merely 
stimulated to germinate by being so boiled (though of course 
the actual germination only takes place subsequently, at lower 

The responses of plants to night temperatures have recently been 
demonstrated to have considerable significance in connection with 
their geographical distribution. Thus the Big Bluegrass (Poa 
ampla) of western North America flowers equally well at day tempera- 
tures of 20, 23, and 30 C. — but only when the night temperature 
is below 14° C, for at 17° C. right temperature it does not flower 


at all, even though vegetative development is good. Again, the 
English Daisy {Bellis perennis) dies when grown continuously in a 
warm greenhouse : with a day-temperature of 26 C. the plants 
survive only at night temperatures below 10 C, and flower 
abundantly only at still lower night temperatures. Accordingly 
such plants are unable to reproduce normally in consistently warm 
climates. In other instances, closely related strains may differ 
markedly in their night-temperature requirements for flowering ; 
these requirements may be decisive in determining which strains, 
if any, can flourish in a particular area. This is true of Tomatoes, 
where fruit-set is dependent upon a very narrow range of tempera- 
tures — a phenomenon which is reflected in very large differences in 
yield in varying circumstances and with different strains having 
even slight deviations in optimal requirements. Numerous instances 
are now known in which, for these or other reasons, slight differences 
in the temperature-response of plants will exert a controlling influence 
on their local survival and consequently on their distribution. 

In the many parts of the world that have markedly varying seasons, 
one of the main concerns of their plants is to tide over unfavourable 
periods — usually of cold or drought. To this end is expended a 
good deal of what might be called evolutionary ingenuity, and also 
much physiological effort — for example, in storing food for the 
adverse period and for subsequent development. Among the most 
successful methods employed is the annual habit, in which the 
adverse period is evaded by being passed over in the form of a 
resistant seed or fruit, the parent having meanwhile died. Numerous 
common weeds, such as Chickweeds (Stellaria spp.) and Shepherd's- 
purse (Capsella bursa-pastoris), practise this method, as do many of 
the diminutive ' ephemerals ' which blossom so pleasingly after rain 
in the less extreme deserts. In the Arctic and some other rigorous 
regions, however, the growing-season, though fairly regular, is too 
short and cool to allow full development — from seed through seedling 
and adult to flower and seed again — in a single season. Accordingly 
almost all the plants there are perennial, passing the adverse winter 
period in a more or less resistant and dormant state — often after 
dying down (in the case of herbs) or losing their leaves (in the case 
of deciduous shrubs and trees). Most Mosses and some other plants 
have the fortunate capacity to endure drought by drying up almost 
entirely without ill effect, resuming normal life again when moistened. 
All these, as well as any growth-responses they involve, are physio- 
logical activities (or inactivities) and, in a sense, adaptations to 


environmental conditions. On a plant's capacity for them may 
depend its geographical range. 

The production of reproductive bodies involves various physio- 
logical activities that are closely correlated and indeed wonderfully 
integrated, yet may be affected by environmental conditions in a 
unique way. Although some trees may live for more than 2,000 
vears, there is no known instance of life being really permanent in 
anv individual. So in order to persist a plant must reproduce, and 
any condition which prevents it from doing so in a particular area 
will preclude that area from its normal range (that is, in the absence 
of persistent immigration). Many conditions — climatic, nutritional 
or otherwise — can and frequently do prevent the normal reproduction 
of certain plants, so limiting the geographical areas they occupy. 
Some plants circumvent this either by separating off parts of their 
bodies for ' vegetative ' reproduction or by the development of 
special organs for the same purpose, thereby enabling themselves 
to persist in areas where seed etc. cannot be produced. This is 
true of many plants living under extreme conditions, for example 
in the Arctic. 

Although on land there is almost everywhere sufficient light to 
enable plants to grow, the effect of light-climate on their reproductive 
processes affords another instance of range-limitation. For many 
plants require a day-length within particular limits before they can 
flower successfully, and, in latitudes where the length of day during 
their flowering period is outside these limits, are unable to reproduce 
sexually. This appears to be one reason why many southern 
species fail to flower in the north, and vice versa. However, such 
reactions are by no means immutable, but tend to vary with other 
conditions, and may also be changed by treatment with certain 

The ranges of particular species may be limited by chemical 
' antagonism ' (i.e. active opposition to growth, etc.), nutritional 
conditions, and other factors bound up with the soil. Familiar 
instances are afforded by some plants which require much ' lime ' 
(actually, calcium carbonate) in the soil, and others which avoid it. 
Examples of the former category are Yellow Mountain Saxifrage 
(Saxifraga aizoides agg.) and Salad Burnet (Poterium sanguisorba), 
and, of the latter, most Heaths (Ericaceae). Often the merest trace 
of a particular compound or element, such as boron, can have a 
profound effect in encouraging or precluding particular species. 
Deficiency diseases, due to lack or insufficiency of particular 


substances, are common. These diseases, -with the ones produced by 
attacks of various Fungi, Bacteria, Viruses, and Nematode Worms, 
and the browsing of lower animals such as Locusts and of higher 
ones such as Goats, may drastically limit plant distributions. Often 
the very presence of a plant species in a spot is dependent upon the 
absence there of serious pests and predators. 

The areas of parasites and saprophytes are naturally limited to 
ones where suitable hosts or elaborated materials, respectively, are 
available for attack. Thus, for example, the devastating Late- 
blight of Potatoes and Tomatoes, caused by the Fungus Phytophthora 
infestans, is limited to the areas of those crops and of some other 
members of the family (Solanaceae) to which they belong. Again, 
the deadly White Pine Blister-rust, Cronartium ribicola, is virtually 
limited to areas supporting both of the hosts that are necessary for 
the completion of its life-cycle, namely, five-needled Pines and 
species of Ribes (Currants and Gooseberries). 

As regards physiological antagonisms due to poisonous residues 
and excretions, it seems that these may be important in some circum- 
stances, such as ' fairy rings ' and the avoidance by some plants of 
the shade of certain trees. Thus the roots of Black Walnut (Juglans 
nigra) have long been known to excrete a toxic substance, juglone, 
that inhibits the growth of many other plants and can even kill 
Apple trees. There are also the cases of the western North American 
members of the Daisy family (Compositae), Parthenium argentatum 
and Encelia farinosa y which are known to poison other plants by 
minute amounts of chemical excretions, thereby reducing competi- 
tion. Is it possible that this may be one of the factors lying behind 
the notorious success of this family as colonists ? We do not know, 
and indeed our information in such fields of study is still only 
fragmentary. Also undetermined but pregnant with possibilities for 
research, is the extent to which antibiotic substances may be effective 
in nature. 

Ecological Limitation 

The realms of physiology range imperceptiblv into those of 
ecology, which in part may be looked upon as the application of 
physiology to ' field ' conditions. The ecological requirements of 
different plants are widely various and, as we have seen, commonly 
limit their geographical areas. This limitation is actually to those 
regions where appropriate ' habitats ' {i.e. living places) exhibiting 


suitable conditions are found, and, within such regions, naturally to 
those habitats themselves. Consequently plants in nature are limited 
not onlv to areas of particular climate but more precisely to special 
habitats within these areas, the final limitation being ecological. In 
such cases as oases in a desert or islands in an ocean, this limitation 
may be extreme. 

The subject of modification by, or adaptation to, various conditions 
is dealt with in the next section. Here we should mention the 
manner in which, quite apart from any special antagonism, sheer 
physical, competition among plants for the requisites of life may 
limit the habitat and actual range of a particular species or even 
strain. Especially may root-competition for water and aerial com- 
petition for light prove veritable struggles for existence in which 
the weaker individuals succumb. Generally speaking, the closer any 
two types are in their ecological requirements, the keener will be 
the competition between them : in such even contests the slightest 
advantage to one competitor can swing the balance in its favour. 

As most plants living on land need soil in which to root and some 
well-lit space in which to grow, it is particularly in ' open ' areas 
not yet covered with higher vegetation that competition is least and 
plants can enter and establish themselves successfully. Most such 
areas tend to be colonized by successive waves of plants that usually 
start with primitive or other lowly types but in favourable regions 
normally lead up to forest. This progressive colonization is called 
1 succession ', and is described in Chapter XI. The farther it 
proceeds, the less space there is left for new colonists and the more 
tendency there is for former colonists to be ousted by coarser 
competitors. Meanwhile animals, including Man, are continually 
opening up new habitats and abandoning old ones — often after 
destroying the natural vegetation, and rarely without disturbing it. 
For these and other reasons the geographical areas of plants and 
plant communities are rarely if ever static. 

Structural ' Adaptations ' of Vegetative Parts 

Numerous features help plants to offset the effects of unfavourable 
conditions and consequently widen their potential ranges. Having 
noted already such functional modifications as acclimatization of 
various sorts, and physiological ' adaptations ' such as the ability of 
many plants to evade unfavourable periods of cold or drought, we 
shall deal here with changes of form that appear to be developed in 


relation to the needs of plants to combat adverse conditions. 
Through such structural changes they may be enabled to maintain 
their geographical areas and even to extend them. Those modifica- 
tions of reproductive bodies that are helpful in dispersal will be 
dealt with in the next chapter, the present section being concerned 
primarily with the ' vegetative ' parts — comprising, in higher plants, 
the stems, roots, and leaves. 

The water relationships of plants often involve strikingly ' adap- 
tive ' features — particularly ones that are helpful in tiding over 
periods of water deficiency, for example by increasing absorption 
or decreasing loss, or by storage against times of need. Instances 
are seen in the deep roots of many plants of deserts or semi-deserts, 
allowing the tapping of underground reserves, and in the matted 
turf of the Grasses of semi-arid regions, which aids retention of such 
water as becomes available from atmospheric sources. Actually, as 
pointed out by Professor Kenneth V. Thimann (in litt.), * roots 
elongate when aerated ; hence in dry soils (which are therefore full 
of air) they grow longer. ... I should call [this] a simple response 
to external conditions. Low nitrogen also favors elongation of 
roots, with obvious ecological advantages in nitrogen-poor soil.' 

The aerial parts of a wide range of plants are modified to reduce 
water-loss, often to the slightest proportions in times of shortage. 
This may be done, for example, by protection of the stomata in grooves 
or among a mass of hairs, by general reduction of the ' evaporating 
surface ', and by covering with wax or hairs, etc., even those areas 
that remain. Often the leaves are reduced to spines or scales, their 
normal functions being taken over by green stems. In addition 
many plants, such as the more massive succulents of the Cactus, 
Spurge, and some other families, store water extensively in special 
stem or other structures which are modified into reservoirs. There 
may also be one or more layers of large water-storing cells in leaves 
and other green parts. The development of some of these features, 
such as the tall stems of many trees in dense forests, may depend 
upon the conditions under which an individual grows, whereas in 
other cases the features may develop regularly, irrespective of local 
conditions, as part of the normal form of the plant. But in either 
instance the ' ability ' has to be present, else the plant could not 
develop the desirable adaptation and benefit accordingly. Fig. 20 
shows some examples of features that help land plants to conserve 
or obtain water ; conversely, many water plants have special tissues 
or growths that enable them to float or otherwise improve their 

(See p. 85.) 


8 4 








Fig. 20. — Features aiding water conservation or absorption. A, branch of a 
desert plant, Hakea, with the leaves modified as spines (X ^); B, stems cf 
Euphorbia tirucalli, specialized for photosynthesis and water storage ( X about \) ; 
C, Arizona desert with large Cacti (phot. F. Shreve); D, cut bank on Jornada 
Experimental Range, New Mexico, showing deep rooting of low desert plants — 
in particular a Mesquite bush about 12 inches (30 cm.) in diameter and only 
6 inches high but with roots about 8 feet (nearly z\ metres) deep (phot. U.S. 
Forest Service) ; E, ' bisect ' diagram of above- and below-ground parts of forbs 
and Grasses in the Palouse prairie grassland association of western central Idaho, 
U.S.A. (courtesy of U.S. Soil Conservation Service) ; F, BlackGrama Grass 
(Bouteloua eriopoda) grown under three degrees of grazing, showing effect on root 
system (courtesy of U.S. Soil Conservation Service). 




aeration, three examples being shown in' Fig. 21. Among these 
last the Water-hyacinth affords an example of how floating may 
aid in dispersal without involving special reproductive bodies, for 
individuals may be transported considerable distances by water cur- 





Fig. 21. — Features promoting aeration. A, Jussiaea repens, a rooting or floating 
aquatic with numerous inflated roots which project upwards into the air and 
contain a great development of air spaces through which air can pass to submerged 
organs (X J); B, Water-hyacinth (Eichhornia crassipes), with leaf-stalks modified 
for buoyancy, the whole plant floating freely (x |); C, cross section of leaf- 
stalk of a Water-lily (Nymphaea stellata), showing large air-passages ( X 30). 

rents, and, having so migrated, often multiply to cover large areas 
of water. 

Also significant in enabling plants to grow in many situations 
where otherwise they could not exist, are modifications for climbing, 
twining, scrambling, and running. Examples are shown in Fig. 22. 
Further modifications apparently playing a similar role in plant 
geography include those for catching insects to supplement the food 
supply, and those for storing food to tide over the adverse period 
of winter. Examples of carnivorous plants and of food-storage in 
special underground organs are shown in Fig. 23 ; included in the 
latter category are Potatoes and many bulbs and other structures 
that are, besides, reproductive in function. 

The giving off of water-vapour from the aerial parts of plants 
helps to keep them cool, and many are further protected from 
intense sunlight by their structure or covering, so that ' scalding ' 
and other injury may be averted even in very hot and sunny deserts. 
The structural changes which restrict or accelerate the rate of water 
loss are in general either hereditary and consequently characteristic 
of the race, or are acquired by an individual plant or part of a plant 





8 9 

Fig. 22. — -Various adaptations for climbing, twining, scrambling, and running. 
A ( X i), leaf-tendrils of Common Pea (Pisum sativum, left) and Clematis (Clematis 
sp., right); B, branches of Bougainvillaea modified as spines used in scrambling 
( f); C, Dodder (Cuscuta), a parasitic twiner that sends haustoria into the host- 
plant (X 1); D, a ' Walking ' Fern (Adiantum caudatum) (X i); E, Ivy (Hedera), 
showing climbing roots ( X ^). 

in response to the particular conditions under which it has grown. 
Thus in the latter instance we may even get large but thin ' shade ' 
leaves and small but thick ' sun ' leaves on the selfsame branch of 
a tree, whereas no matter under what conditions most compact 
desert plants are grown they will not become tall and lax, the char- 
acteristic of compactness being in such instances usually hereditary 

9 o 





Fig. 23. — Modifications for storing food or catching insects. A, expanded 
storage-root of Turnip (Brassica campestris) (X |); B, Ginger {Zingiber) plant 
with enlarged storage rhizomes ( X £); C, bulbs of Lily (Lilium sp., left) and Onion 
(Allium cepa, right) (X f); D, Sarracenia, a Pitcher-plant, showing flowers and 
pitcher leaves (x %); E, Sundew (Drosera), a carnivorous plant (X |). 

and ' fixed ' through long evolutionary history. 1 Of such a deep- 
seated and lasting nature are most of the vegetative and reproductive 
features which go to make up a plant species, giving it its special 
form or morphology. By this we classify it as part of a systematic 
hierarchy in the manner explained at the beginning of Chapter II. 

1 Often the same character-manifestation is hereditary in one group of plants 
and due to direct environmental impress in another — an example of the latter 
being the compact form of many alpine plants as opposed to those characteristic 
of deserts. In such instances special cultivation may be necessary to determine 
to which category a feature belongs. 

92 introduction to plant geography [chap. 

Classification by Life-forms 

The ' life-form ' or ' growth-form ' of a plant is the form which 
its vegetative body produces as a result of all the life-processes, 
including those that are affected by the environment within the 
plant's life-time and are not heritable. Although a plant's life-form 
is among its most striking characteristics, it may be of a rather fickle 
nature. Thus different individuals of the same species can some- 
times belong to different life-forms, for example when they have 
been grown in different environments ; for under any particular 
life-form are merely grouped together those plants which, in their 
entirety, show similar morphological adjustments. Life-forms may 
accordingly give a fair indication of environmental impress, or at 
least tell us something about local conditions. 

Although the description of vegetation in terms of life-forms is 
widely imprecise, and classification by them is inadequate for our 
ultimate purpose, nevertheless it is a part of common parlance and 
can be of some value. Its use goes back at least to the times of 
the ancient Greeks, who classified plants into trees, shrubs, herbs, 
etc., which are among the most obviously differing life-forms. Even 
nowadays to the general geographer or other non-biologist the species, 
etc., making up plant communities are often less significant than the 
prevalent life-forms. These last may yet be of importance in two 
allied biological fields, namely, plant sociology, where consideration 
of life-forms may help in the description of the structure of the 
communities that are the main subject of study, and ecology, where 
mention of the predominant life-form is often sufficient to give some 
idea of the local environment. 

In spite of the limitations mentioned above, there is one particular 
system of life-forms which as plant geographers we may find useful, 
although it suffers from rather difficult Greek terminology. As 
originally elaborated by the late Professor C. Raunkiaer of Denmark 
and usefully modified by, among others, Dr. J. Braun-Blanquet of 
Montpellier, this system lays stress primarily on the adjustment of 
the plant to the unfavourable season, and particularly employs the 
position of the perennating 1 buds relative to the soil surface in 
attempting to classify together plants of similar habit. The result is a 
series of life-forms that is especially interesting to the more statistic- 
ally minded among us, the main categories of which are as follows : 

1 Perennation is the act of tiding over an unfavourable period, such as a cold 
winter or a dry summer. 


(a) Phanerophytes (tall aerial plants). Perennials, mostly trees or 
shrubs, with their renewal buds on shoots at least 25 cm. (about 
10 inches) above the surface of the ground, and hence exposed to 
unfavourable weather. Phanerophytes are especially numerous in 
moist areas of the tropics and subtropics, where they tend to pre- 
dominate in the matter of numbers of species as well as individuals. 
Elsewhere the species are usually few, even if the numbers of 
individuals are great and their dominance is overwhelming. 

(b) Chamaephytes (surface plants). Perennial herbs and some 
undershrubs with renewal buds between ground-level and a height 
of 25 cm. — hence usually enjoying only such protection as may be 
afforded by the plant itself or by snow, and consequently plentiful 
in boreal and alpine regions. 

(c) Hemicryptophytes (half-earth plants). These have perennial 
shoots and buds at ground-level or within the surface layer of soil, 
etc., and hence protected by the habitat. Such plants are particularly 
preponderant in high alpine and arctic regions but are also plentiful 
in the temperate zone. 

(d) Geophytes (earth plants). These have the perennating organs 
(such as bulbs, tubers, or rhizomes) well buried in the soil and there- 
fore not exposed in unfavourable seasons. They tend to be com- 
monest in temperate regions but also persist in fair numbers farther 
north and south. 

(e) Hydrophytes (water plants). These include all water plants, 
whether anchored or not, apart from microscopic free-floating or 
swimming types which form the main basis of the separate category 
known as ' plankton '. This group of hydrophytes tends to cut 
across the other main ones and so is often omitted from ' spectra ' 
(see pp. 94-5). 

(/) Therophytes (annuals). Plants which complete their life-cycle, 
from germination to ripe seed, within a single limited vegetative 
period, surviving the unfavourable times as seeds, spores, or other 
special (usually resistant) reproductive bodies. They are especially 
abundant in deserts where the unfavourable period may be par- 
ticularly severe and prolonged, but are largely lacking in arctic 
regions where the growing-season is too short or the warmth is 
insufficient to allow them to complete development before winter 
comes again. 

Examples of (b), (c), (d) and (/) are illustrated in Fig. 24. Almost 
all trees and tall shrubs belong to (a), while examples of (e) were 
illustrated in Fig. 21. 




Further categories may, if desired, be added to the above system 
— such as epiphytes growing on trees etc. Moreover, refinements 
may be used such as the subdivision of phanerophytes into nano- 
phanerophytes (shrubs) in which the renewal buds lie less than 2 metres 
above ground, microphanerophytes (small trees) in which they lie 
at a height of from 2 to 8 metres, the taller mesophanerophytes 
(8-30 metres), and the still taller megaphanerophytes (above 30 metres) ; 
also phanerophyta scandentia (lianes) which are woody climbing 
plants whose renewal buds pass the unfavourable season high above 
the ground. 

Fig. 24. — Diagrams illustrating some Raunkiaer life-forms. A, a creeping 
chamaephyte; B, a rosette hemicryptophy te ; C, a tufted hemicryptophyte ; D, 
a bulb geophyte; E, a rhizome geophyte; F, a therophyte. (After Braun- 


The values of this system are relative, its applications limited. 
Being based on wide life-form categories, it certainly cannot take 
the place of detailed description of vegetation including naming of 
the main species concerned, which alone will indicate to the qualified 
reader the precise nature of each named species and, through these, 
reveal much concerning the community itself. Its use is, moreover, 
limited in arctic and alpine regions where the success of a particular 
plant in life is apt to depend not so much on its adaptation to a 
rigorous winter as on its adjustment to the very short and cool 
summer. Nevertheless, in the hands of the student who is statistic- 
ally but perhaps not taxonomically minded and trained, not wanting 
or able to name specifically the plants concerned, this system is 
useful in giving a fair analysis of the components of a community 
or flora in terms of the representation of each life-form. 

Such an analysis is usually expressed as a ' biological spectrum ', 
indicating the percentage of the total flora belonging to each of the 
life-forms involved. Considering only vascular plants and excluding 


hydrophytes, examples from areas in the main climatic belts whose 
land-vegetation is described in Chapters XII-XIV are as follows, 
in round figures : 

(i) Temperate — phanerophytes 15, chamaephytes 2, hemicrypto- 
phytes 49, geophytes 22, therophytes 12 ; 

(ii) Arctic — phanerophytes 1, chamaephytes 22, hemicryptophytes 
61, geophytes 15, therophytes 1 ; 

(iii) Tropical (moist) — phanerophytes 61, chamaephytes 6, hemi- 
cryptophytes 12, geophytes 5, therophytes 16 ; 

(iv) Tropical (arid) — phanerophytes 9, chamaephytes 14, hemi- 
cryptophytes 19, geophytes 8, therophytes 50. 

With the above it is interesting to compare the ' normal ' spectrum 
for the world as a whole, which is claimed to be : phanerophytes 
46, chamaephytes 9, hemicryptophytes 26, geophytes 6, therophytes 


Altogether it may be concluded that such life-form spectra can 

give a useful if generalized impression of the biological effects of 
climatic features and hence help characterize the various phytogeo- 
graphical regions. But dealing as they do with wide categories, 
and with flora rather than vegetation (that is, with the different 
kinds of plants inhabiting an area regardless of their abundance and 
relative importance), they are no adequate substitute for more 
thorough description with structural details, precise naming, and, 
wherever possible, good illustration. Thus, for example, a small 
group of species or even a single species may dominate and largely 
characterize a plant community or sometimes a whole region, and 
yet scarcely ' tell ' in the spectrum. This, however, is an objection 
to the spectrum method of counting species rather than to the life- 
form classification itself. 

Further Consideration 

The principles of plant physiology can readily be acquired from W. O. 
James's An Introduction to Plant Physiology, fifth edition (Clarendon 
Press, Oxford, viii -f- 303, 1955) or, in more detail, from such a text as 
B. S. Meyer & D. B. Anderson's Plant Physiology, second edition (Van 
Nostrand, New York etc., pp. viii -f- 784, 1952). 

More details and examples of structural ' adaptations ' that apparently 
enable plants to maintain or extend their geographical ranges, may be 
gained from almost any good modern work on structural botany, or from 
G. Haberlandt's classic Physiological Plant Anatomy, translated by M. 
Drummond (Macmillan, London, pp. xv + 777, I 9 I 4, reprinted 1928). 


The system of life-forms outlined above is clearly elaborated in Chapter 
XII of J. Braun-Blanquet's Plant Sociology (McGraw-Hill, New York & 
London, pp. xviii -f- 439, 1932) ; there are some refinements in the second 
German edition, Pflanzensoziologie : Grundzuge der V egetationskunde 
(Springer, Wien, pp. xi -f 631, 1951). However, for a detailed account 
of the development and application of this system, the interested student 
should refer to the volume of collected papers of the late C. Raunkiaer, 
entitled The Life Forms of Plants and Statistical Plant Geography (Claren- 
don Press, Oxford, pp. xvi -f 632, 1934). A briefer account is given in 
the same author's Plant Life Forms, translated by H. Gilbert-Carter 
(Clarendon Press, Oxford, pp. vii -f io 4, 1937). 

Any walk in the country, or even in a garden or public park, with due 
contemplation of the seemingly endless variety of plants encountered — 
the Lichens or green powdery algal cells on the bark of many trees are 
just as truly plants as the giants on which they grow — should convince 
even the most sceptical layman of the need for classification. The more 
intelligent and interested will almost inevitably find themselves comparing 
similar plants and mentally putting them into groups, which may be 
either systematic or life-form ones. It may be noted in the course of 
such observations that the life-forms chiefly give some indication of the 
physiognomy of the vegetation. This is largely dependent on local 
environmental conditions and may look alike even where quite different 
kinds of plants are involved. On the other hand, systematic relationships 
(e.g. following the lines indicated in Chapter II) depend also considerably 
on past and present geographical connections and barriers, so that only 
an account including floristic determinations and details of frequency 
etc. can give the more complete picture for which we strive. 

Chapter IV 


Having stated our objectives and familiarized ourselves with the 
main groups of plants, we must consider the methods by which 
different plants increase their areas, at least potentially, by special 
' adaptations ' of the reproductive bodies and by seizing such 
opportunities for their transport as may be offered. These adapta- 
tions are of the nature of beneficial structural modifications (see 
Chapter III). The areas attained are the mainstay of our plant 
geographical studies, and although they are liable to be profoundly 
affected by past history (as we shall see in the next two chapters) 
and are further greatly limited by the physiology of the plants 
themselves (as we have already seen in Chapter III), these areas 
must to a large extent be a function of the plants' own aptitudes. 
In the final analysis, areal spreading is often limited by the ecological 
reactions of the plant to a new environment which may, for example, 
be too cold or too dry for its successful establishment. Such 
reactions are primarily physiological, and, though their outcome is 
often capable of modification, as we have already seen, they com- 
monly determine the potential or ultimate area which a species can 
occupy when there is fully effective dispersal. The actual areas 
within the physiologically circumscribed potential ones are largely 
determined by barriers to successful migration. 

It should be noted that dispersal and migration, although closely 
connected, are different activities. Dispersal merely involves_dis- 
semination from the parent and distribution (in the dynamic sense) 
to a new spot, whereas migration implies also successful growth and 
establishment (ecesis). Thus dispersal is a necessary forerunner of 
migration, which is actually accomplished only on establishment in 
a new place. In nature only a small proportion of the plant bodies 
which become dispersed, and which may conveniently be termed 
disseminules (dias.po.res), actually become established and effect 
migration. Not only do many of them die prematurely or fall on 



* barren ground ', or come to rest where they cannot even start a 
new life, or fail to survive the struggle with stronger competitors, but 
the ecological conditions and physiological reactions have to lie 
within often narrow limits for ultimate success. In any case, there- 
fore, the vast majority of disseminules are doomed. 

These disseminules, the actual bodies moved, are most often 
reproductive structures such as spores, seeds, or fruits. In numer- 
ous instances, however, they are special structures of a vegetative 
nature, or unmodified parts of plants, whole plants, or even groups 
of plants — though in the last instance usually effective only by 
chance. An example of a whole plant being transported was the 
Water-hyacinth mentioned in the last chapter and shown in Fig. 21 , B. 

Often the same plant species or individual will produce more than 
one type of disseminule, thereby increasing its chances of effective 
migration. Thus, whereas the majority of our familiar north- 
temperate forest trees, such as Oaks and Spruces, normally repro- 
duce by seed, they may also do so by means of suckering, layering, 
or other vegetative activity. Moreover, many of the plants that 
are most successful in colonizing vast areas, resort to more than one 
means of dispersal. Thus the Common Reed (Phragmites com- 
munis agg.), which is often claimed to be the most widely distributed 
vascular plant species in the world, has the multiple advantages of 
a wind-dispersed, plumed fruit, and a water-dispersed, more or less 
buoyant rhizome — besides considerable variability in form, and an 
ability to occupy a wide range of moist to aquatic habitats. These 
it colonizes so aggressively and holds so strongly that its ' beds ' 
form a formidable barrier against immigration by other plants. On 
the other hand, one of the numerous unsolved problems of plant 
geography is that of why many plants with seemingly excellent 
advantages in dispersal are not widely distributed. Yet another 
major question is posed by the number of groups and even species 
that are widespread without seeming to have any adequate means 
of dispersal. That precisely the same type of plant should have 
evolved separately in several different places is almost unthinkable 
to most students, and so it is widely assumed that the areas currently 
occupied by particular plants are due to dispersal and effective 
migration. Now that we have explained the distinction between 
these terms, they need not henceforth be separated. Rather will 
we refer to dispersal when the question of establishment can be 
ignored, and to migration when such establishment is to be 

4] dispersal and migration 99 

Wind Dispersal 

A walk in the woods and fields of a north-temperate region on 
a boisterous autumn day should convince any sceptic that air currents 
of one kind or another are important in the dispersal of many 
different plants. Not only do winds blow leaves, and sometimes 
small branches, about — and with them adhering parasites or sapro- 
phytes, for example — but they obviously transport some seeds and 
fruits for considerable distances. The more efficiently adapted of 
these, whose bodies are so light or whose ' form-resistance ' is such 
that they sink only slowly in still air and float almost indefinitely 
in a light breeze, may be transported far from the parent. This 
undoubtedlv happens with such plumed seeds as those of Milkweeds 
(Ascleptas spp.) and Fireweed (Epilobium angustifolium agg.), or with 
such ' parachute ' fruits as those of Dandelions {Taraxacum spp.) 
and many other members of the Daisy family (Compositae). 

Even more effective is the dispersal by air currents — including 
upward eddies that carry them into the upper atmosphere — of 
microscopic spores, especially of Fungi and Bacteria. Such dis- 
persal may take place over distances that in numerous instances 
have been proved to run into many hundreds of miles. The present 
writer has studied this subject for years and is convinced that these 
smaller ' botanical particles ' or ' spora ' can be (and often are) 
carried thousands of miles in the atmosphere, frequently without 
losing their power to resume active life on regaining suitable con- 
ditions. Thus he has trapped some spora in the immediate vicinity 
of the North Pole under both winter and summer conditions, as well 
as elsewhere at vast distances from their nearest conceivable point 
of origin. 

Quite apart from disseminules which are specially modified for 
transportation by winds, and others which are so minute that they 
need not be so modified to be transported, there are many recorded 
instances of large and heavy bodies being blown for considerable 
distances by hurricanes, etc., on special occasions. After the 
devastating tornado in and around Worcester, Massachusetts, in 
June, 1953, abundant shingles and often bulkier roofing materials 
and sizeable living branches of trees were to be seen littering the 
ground fully 20 miles nearer the coast than the closest point at which 
the ' twister ' had struck. There are also records of windfalls of 
uprooted plants scattered over wide areas. It need scarcely be 
remarked that, as successful transportation and growth of a single 


plant or disseminule is sufficient for its establishment in a new region, 
even extremely rare occurrences may be important and involve 
quite unexpected species and circumstances. Instances in point 
include exceptional winds in various regions, and the blowing of 
seeds, fruits, or whole plants over the ice or compacted snow in 
arctic regions in winter. Not only may such blowing over ice be 
effective from time to time, in the case of higher plants, but, more 
often as they tend to remain longer alive, it may also result in the 
dispersal of parasitic or saprophytic Fungi, etc., growing upon or 
within the bodies of these higher plants. When we recall that, not 
very many thousands of years ago, ice covered vast tracts of what 
are now among the most populous parts of the northern hemisphere, 
as well as, doubtless, the adjacent seas, we can imagine that such 
dispersal may have been of great importance in the migrational 
history of plants in areas far south of the present-day Arctic. 

It is instructive to consider briefly the main organs or methods 
of wind-dispersal, and particularly those plant bodies which are 
especially modified for the purpose. For this, there should be 
recalled the distinction between seeds and fruits which was explained 
on page 69, and the very different origin and nature of spores in 
different cases. 

(a) Spores. These, as we saw in Chapter II, are the main dis- 
seminules of most of the groups of plants up to and including the 
Ferns, and are often produced in fantastically great numbers. Thus 
a single specimen of the Pasture Mushroom (Agaricus (Psalliota) 
campestris) has been estimated to produce 1,800,000,000 spores, 
while a large specimen of the Shaggy-mane Mushroom (Copn'mis 
comalus) may produce 5,240,000,000 spores, and some Purr balls 
many times that number ! Although extremely variable in size 
and form, spores are commonly minute and easily blown about by 
the wind — being frequently borne by upward air-eddies rising from 
warm plains and carried into the upper atmosphere where they mav 
be transported vast distances. Indeed, like volcanic dust, they are 
probably sometimes blown around the world without settling to 
earth. Bacterial and some other minute cells may belong to the 
same category as spores in the matter of size and aerial buoyancy. 
The spores are often extremely resistant to low temperatures and 
desiccation which in fact appear to prolong their life, so that many 
caught in the most remote situations are alive, able to germinate 
when given suitable conditions, and, as we say, ' viable '. They 
may live for many years and, apparently, often withstand the radia- 


tion effects of high altitudes. According to Ridley, whose monu- 
mental work on plant dispersal is cited at the end of this chapter, 
' There is no part of the world where some are not present, and 
there appears to be a constant rain of the more minute kinds falling 
everywhere.' With little doubt this easy w T ind- dispersal of many 
of the Bacteria, Fungi, and other so-called spore-plants is the 
primary reason for their extremely widespread distribution ; a 
secondary reason is their often wide tolerance of conditions and 
modest requirements for life. 

(b) Dust seeds (and minute fruits). The seeds of many plants, 
such as the members of the Orchid family (Orchidaceae), and the 
one-seeded fruits (for example) of some of the mainly tropical 
parasitic family Balanophoraceae, are also minute and extremely 
light, as well as sometimes winged, and so tend to be blown away 
and about in much the same manner as spores. 

(c) Plumed seeds. These usually bear a light tuft of silky hairs f OjJj^AyO 
at one end and are liberated from a capsular fruit which, on splitting, 
only releases them gradually, often one by one. The plants involved 
are usually herbs or climbers, good examples being species of 
Willow-herb (Epilobium) and Milkweed (Asclepias), and they gener- 
ally occur in open situations, in or from which they can travel for 
hundreds of miles. 

(d) Plumed fruits. These include the familiar ' parachutes ' of 
Dandelions [Taraxacum spp.), the long feathery fruits of species 
of Avens (Geum), and the silky-haired ones of Cotton-grasses 
{Eriophorum spp.). Their appendages cause them to be detached by 
the wind and floated away, often for very considerable distances. 
The plants concerned are usually herbaceous, and include many 
Grasses. An extreme case is that of some disseminules of Grasses 
which have been trapped in the air several thousands of feet above 
the ground, and in view of the highly fortuitous nature of such 
observation it would seem likely that they may reach the upper 
air currents quite frequently. 

(e) Winged seeds. In these it is usually a thin portion of the 
seed-coat which forms a wing that catches in the wind when they 
are liberated, often in considerable numbers, e.g. by splitting of 
the containing fruit-wall. They chiefly occur on trees, shrubs, 
and lianes (woody climbers), and so are liberated some distance 
above the ground — which is just as well, for their dispersal mechanism 
tends to be much less efficient than those of the categories mentioned 
above. Good examples are afforded by members of the Bignonia 


family (Bignoniaceae), and by Pines and Spruces and many other 

(/) Winged fruits. Again chiefly occurring on trees and shrubs, 
these are so modified as to cause the fruit, on detachment by the 
wind, to be borne at least out of the immediate sphere of influence 
of the parent plant, or to trundle along as is the case with many 
bladder-fruits. Often the flight is a spinning one and, though 
spectacular, not very efficient in terms of distance. Each fruit (as 
n the Birches, Betula spp.), or half of a separating fruit (as in most 
Maples, Acer spp.), is usually one-seeded — functionally, at least. 

(g) Long-haired seeds and fruits. These are sufficiently alike to 
be considered together, while also approaching (c) and (d), their 
main feature being that the surface is covered with long silky or 
woolly hairs. Such disseminules tend to be less efficient than 
plumed ones but are nevertheless capable of travelling for some 
miles. The plants, as in categories (e) and (/), are most commonly 
trees or shrubs. Examples of seeds of this nature are those of 
Cotton {Gossypium), Willows {Salix spp.) and Poplars (Populus 
spp.) ; and of fruits, those of some Anemones (Anemone spp.). 
That this mode of dispersal is abundantly effective, at least so far as 
transport of the seeds is concerned, has been frequently and 
strikingly demonstrated to the writer when he has looked out from 
his laboratory windows in the ancient Botanic Garden at Oxford 
and thought a snow-storm was raging, the ' flakes ' being masses of 
hairy seeds blown from pollarded Willows mostly hundreds of vards 

(h) Tumble-weeds. Such plants, or detached portions bearing 
the seeds, tend to roll before the wind or be blown across open 
country, usually scattering their seeds or fruits as they go. They 
are commonly short-lived herbs that branch densely and stiffly from 
a central stem and have a rounded form. Normally they break off 
easily near ground level and have the seeds or fruits so loose that 
they are lost as the aerial part trundles along. Tumble-weeds occur 
chiefly in deserts or arid prairies or steppes. Examples include the 
so-called Russian-thistle (Salsola pestifer) in North America and 
Eryngium sp. on the northern border of the Sahara in Egypt (R. W. 
Haines voce). In less ideally displayed form, all manner of plants 
or parts of plants can, in special circumstances, act fortuitously as 
tumble-weeds — including the so-called Rose of Jericho (Anastatica 
hierochnntind), and some Lichens and Mosses in the Arctic. 

(/') Other organs or methods. These include pieces of such 


epiphytes (plants growing on other plants) as Spanish-moss (77/- 
landsia usneoides) which get blown to new situations on the trees on 
which thev grow, often abundantly ; seeds which fortuitously stick 
to or get curled up in dead leaves and are transported with them 
for considerable distances ; small seeds or fruits which adhere to 
sticky stalks (for example of Catch-flies, Lychnis spp.) that are blown 
about after detachment ; and soredia of Lichens as well as bulbils, 
for example of such Grasses as Poa alpina, that may be usefully 
scattered by the wind. 

(j) Jactitation. This is the slinging of seeds out of fruits such 
as the capsules of Poppies (Papaver spp.) or Mulleins (Verbascum 
spp.), which are held aloft on long stalks that are liable to be bent 
before the wind or jolted by passing animals — often springing back 
subsequently to jerk out some more of the contents in the opposite 
direction. Such a ' censer mechanism ' is commonly feeble, barely 
(or even not at all) removing the seed from the immediate sphere 
of influence of the parent ; but given the good fortune of a strong 
wind to carry the seed farther, or a favourable slope down which it 
can bounce and roll, jactitation may occasionally be quite effective. 

Fig. 25 shows a wide range of wind-dispersed disseminules. 

Here it seems reasonable to suggest that the primary objective 
of a disseminule, so far as transportation is concerned, is to get 
away from the immediate parental influence and possible competition 
of seedlings developing from its brothers, which for many small 
plants is effected by displacement of merely a matter of centimetres. 
Most dispersal is probably of this relatively minor nature, the long- 
distance ' saltatory ' dispersal (that may drastically extend the area 
and ultimately increase the importance of a race) being supposedly 
much rarer. 

Before proceeding to the next main topic we should give some 
consideration to the barriers and deterrents to wind dispersal, 
remembering that it often includes blowing about on the surface of 
water whose currents may, moreover, cany originally airborne 
disseminules much farther. Wind dispersal operates chiefly on free, 
air-buoyant spores etc. in open places—so that it is not unexpected 
to find that in treeless, high-alpine and arctic regions an unusually 
large proportion of the native plants have wind-borne disseminules, 
whereas in dense forests and other sheltered areas wind is little 
effective. A great deal of wind dispersal, at least of the larger fruits 
and seeds, is discontinuous, bodies being blown up by a gust of 
wind and soon alighting to await another gust, the process in some 






Fig. 25. — Wind-dispersal mechanisms and disseminules. A, capsule of an 
Orchid (Cymbidium), open, with minute seeds being scattered by the wind (X ^); 
B, fruit of Milkweed (Asclepias), showing liberation of the effectively plumed 
seeds ( < ^); C, ' parachute ' fruit of Dandelion (Taraxacum) ( X 2); D, flattened 
seed of Macrozanonia with large papery wing (x "A") J E, pollen grain of a Pine 
(Pinus), with inflated, bladder-like 'wings' making it buoyant in air (x 390); 
F, flattened fruits of an Ash (Fraxinus) (X ii); G, fruit of Maple (Acer) with 
flattened wings (,< 1); H, fruit of Linden (Lime-tree, Tilia), adapted for wind 
dispersal by being attached to a specialized leaf (bract) ( X f); I, capsule of Poppy 
(Papaver), from which the seeds are liberated only on violent shaking ( X 2%). 

instances being repeated again and again. When such bodies alight 
on even small tracts of water, these are apt to constitute insuperable 
barriers to disseminules which cannot float for a protracted period. 
Thus it has been observed that plants depending on winged seeds 
or fruits for their dispersal are rare on oceanic islands. Dense 
forests may have an effect similar to oceans, though of a less finite 
nature. However, the fact that many disseminules await a parti- 
cularly strong gust of wind before becoming detached from the parent, 
is obviously advantageous in that such stronger winds are the more 
likely to carry them afar. 

Mountain ranges also prove a barrier in many instances — though 
the lighter disseminules are easily blown up and over them — as, 
to a lesser degree, do cliffs, walls, and fences. This is evidenced 
by the fact that beneath such obstacles a wide range of wind-borne 
seeds and fruits are often to be found germinating, having been 
stopped in their flight and fallen down. Pits and other depressions 
have much the same effect in providing a barrier against the heavier 


disseminules. The lightest and most effective of these bodies, on 
the other hand, come to grief chiefly through the action of moisture 
which clogs their ' flying apparatus ', or condenses on them and 
weighs them down. In this connection rain is extraordinarily 
effective in removing, often during a single shower, practically all 
of even the lightest spores, pollen grains, etc., from the atmosphere 
through which it falls. For this reason, and because the strongest 
and most lasting winds are chiefly at high altitudes, it is mainly 
those botanical particles which reach the upper air which travel 
really great distances. The fact that many do so appears to be 
primarily due to the upward air currents resultant on the warming 
of dark' land-surfaces by radiant energy absorbed from sunlight. 
Finally, for effective migration the plant has to become established, 
and to that requisite any lack of suitable climatic or edaphic or 
other conditions constitutes an insuperable barrier. 

Dispersal by Water and Ice 

The earliest forms of plant life were probably aquatic and water- 
dispersed, and water still plays a very important part in the dispersal 
of plants — particularly of those that live in or near it. But although 
modifications that appear to be for water dispersal are found in a 
wide variety of land plants, they are not so striking, or so widely 
necessary, as those for wind dispersal. For practically any light 
disseminule may be effectively dispersed by water up to the limit 
of its ability to float and retain the power of germination — that is, 
until it becomes waterlogged and sinks or decays, or until it is 
killed, or, having begun to germinate, has failed to reach a suitable 
habitat. Hence the main requirements for water dispersal are 
sufficient buoyancy and impermeability, their degree of development 
in a particular disseminule being often the most important factor 
determining its success. 

Among Algae and many higher plants (such as the Canadian 
Water-weed, Elodea canadensis) which normally live submerged in 
water, there is no need for impermeability : the plant or special 
disseminule merely drifts with any water current, sometimes attached 
to floating logs, etc. Such drifting appears to be the main mode 
of distribution of most seaweeds. Free-floating plants such as 
Duckweeds (Lemna spp.) or Water Crowfoots are widely dispersed 
as they float on the surface of the water, though they may sink to 
the bottom to perennate. A fine tropical example is the Water- 


hyacinth {Eichhornia crassipes), whose dilated leaf-stalks act as floats, 
as illustrated in Fig. 21, B. It is not, however, by any means neces- 
sary to float on or drift in the body of water to be water-dispersed. 
Thus some seeds or fruits sink at first but rise to the surface on 
germination, to drift until they become stranded — perhaps under 
conditions ideal for further growth — while many are carried short 
distances by rainwash or sudden rushes of water over the ground 
or frozen surface, for example during snow-melt in alpine and arctic 
regions. Severe floods may dislodge and transport whole trees, 
as well as innumerable seeds and fruits that are deposited on the 
wet flood-plain when the water ultimately recedes. Also apt to 
transport living materials are islands of drifting branches etc., ice- 
bergs, drifting ice-floes, and the still larger and more lasting ice- 
islands. These are largely fortuitous and probably capable of 
involving almost any category of plant from time to time, whereas 
the regularly water-dispersed plants normally live in or near the 
water and are modified accordingly. 

The main modes of water dispersal may now be considered : 

(a) Sea currents. These can cause very effective long-distance 
dispersal of suitably modified disseminules, in some known cases 
for over 1,000 miles. For this the body must normally be able to 
float for a long time without becoming waterlogged and must also 
belong to a littoral species that can establish itself under saline con- 
ditions on a sandy, muddy, or other sea-shore. Coconuts are sc 
dispersed, even if there is some doubt as to whether actual migration 
is thereby effected ; and among familiar plants of north-temperate 
and boreal shores that evidently migrate in this manner may be 
mentioned the Oysterleaf (Mertensia maritima agg.) and Sea-beach 
Sandwort (Arenaria (Honckenya) peploides agg.). Excellent tropical 
examples are afforded by the characteristic dominants of mangrove 
swamps, such as species of Rhizophora and Avicennia, the seedlings 
of which float widely. Also normally dispersed by sea currents in 
the manner of seaweeds are further herbaceous maritime Angio- 
sperms — most often as whole plants, parts of plants, or asexual 
propagules. On the other hand, the vast numbers of seeds and 
fruits of freshwater and normal land plants, and of course individuals 
themselves, that are blown into the sea or carried thereto by rivers, 
in general perish. 

(b) Rivers and streams. These commonly transport fruits, seeds, 
and other parts of plants — sometimes as far as from their sources 
right down to the sea. In other cases they may help with the seeding 


of inundated areas. Such dispersal is, however, virtually limited 
to the direction of the current and to the particular land-mass con- 
cerned, the disseminules of other than marine and strand plants 
rarely surviving protracted flotation in the ocean. Thus all manner 
of seeds, fruits, and living fragments of aquatic or river-bank plants 
are to be seen among the ' flotsam ' of debris floating downstream, 
often to be left stranded in situations suitable for growth and 
establishment, while in tidal estuaries migration is often away from 
the mouth of a river, aided by tides which run upstream as well as 
down. The ebb of a high tide where the water is brimming widely 
is particularly effective in the deposition of floating materials. 
Examples of flowering plants regularly dispersed by freshwater 
streams are many of the Pondweeds (Potamogeton spp.) whose small 
fruits in some instances can float for months on end, and the Yellow 
Water-lily (Nuphar lutea), the pulpy fruit of which floats for a few 
days before disintegrating and releasing the seeds, which sink and 
later germinate. An example of a species whose seeds, as such, are 
commonly distributed by water, is the Summer Snowflake (Leucojum 
aestivum). Casually, almost any plant or its disseminules may be 
transported downstream by flotation, striking examples being the 
alpine species that are often to be found in open streamside habitats 
in the lowlands. Familiar instances in the boreal regions are such 
' open-soil ' types as Mountain Sorrel (Oxyria digyna), Moss 
Campion (Silene acaulis agg.), and various Saxifrages. 

(c) Rainwash, floods, and lakes. Rain not only splashes out the 
seeds or spores from open organs but, wh en forming a wash, may 
carry them much farther than other agencies commonly do — 
especially when it develops into a flush or extensive run-off, perhaps 
in time to form a rivulet or even to join a major stream. A con- 
siderable run-off may be noted in boreal regions when_the_ snow 
melts in spring but the ground remains frozen and impervious. 
Often it is not necessary that disseminules, in order to be washed 
away, should be able to float, though to reap the benefit of wider 
dispersal by ordinary floods they should do so, as of course they must 
normally do to be blown about on lakes. Almost any plant or part 
of a plant may in certain circumstances be dispersed by drastic 
floods, involving as they do the uprooting of trees and the carriage 
of all manner of debris, sometimes for considerable distances — per- 
haps to be deposited in a silty flood-plain well suited to the establish- 
ment of migrant plants. In lakes the methods and plants involved 
are in general similar to those in streams, but there is more limitation 


of effective dispersal to aquatic and semi-aquatic types, and the 
distances of dispersal are usually small. Most often, partially corky 
or other air-containing tissues cause the body involved to float, or 
buoyant vegetative parts are detached by feeding Mammals or 

Fig. 26 shows a range of water-dispersed bodies. 

(d) Icebergs, ice-floes, etc. The ' rafting ' of all manner of 
material, including living plants and their disseminules, after blow- 
ing, falling, or spring-time washing on to fast-ice near the shore 
or on to glaciers which later ' calve ' to form icebergs, has been 
widely recognized in arctic and subarctic regions. There can be no 
doubt that by this means much material is transported out to sea 
and often far away before the ice melts and releases it, though it 
seems unlikely that the disseminules of land plants find their way 
back to suitable habitats at all frequently. Probably more important, 
and certainly more frequent, is the dispersal of Diatoms, particularly, 
which grow upon the ice-floes and may in time travel hundreds or 
even thousands of miles with them, or of such strand-plants as 
Creeping Alkali-grass (Puccinellia phryganodes agg.) which are ' picked 
up ' after being frozen solid in ice that forms about the shores on 
which they grow. It may be presumed that these occurrences were 
more widespread during the Ice Ages, though there are instances 
occurring even well south nowadays — e.g. in the estuaries of the 
Atlantic seaboard of the United States. Ice floating down rivers or 
blown about lakes may also be of significance in carrying disseminules 
that do not float. The present writer has investigated the plant 
materials collected on a large ice-island in the vicinity of the North 
Pole, that had drifted many hundreds of miles from the point 
where they were washed or blown down from the land on which 
they grew. Almost all of these materials were dead, but those 
collected when the ice-island had drifted at the very least 3,000 miles, 
and quite possibly several times that distance, included an extensive 
though thin tussock of the Moss Hygrohypnum polare which was 
found to be still alive. 

Charles Darwin, in The Origin of Species (6th edn. 1873, P- 3 2 6)> 
after noting that the natives of the coral islands in the Pacific procure 
stones for their tools solely from the roots of drifted trees, remarked 
that these roots also frequently enclose small parcels of earth ' so 
perfectly that not a particle could be washed away during the longest 
transport : out of one small portion of earth thus completely enclosed 
by the roots of an oak about 50 years old, three dicotyledonous plants 





Fig. 26. — Water-dispersed fruits and other bodies. A, sectional view of fruit 
of Coconut (Cocos nucifera) showing the thick fibrous outer husk which encloses 
much air and enables it to float protractedly (X -A-); B, germinating seedling of 
a Mangrove (Rhizophord) projecting from a fruit that is still attached to the tree 
(many such seedlings on detachment can float in the sea for weeks on end) ( X ',); 
C, inflated capsules of Cardiospermum, the one on the right having been cut through 
to show the contained seeds (such fruits may be blown about as well as float) 
(X f); D, fruit of Heritiera littoralis, adapted for water dispersal by its thick 
fibrous husk enclosing an air-cavity (seen in the half-specimen below) (x £); 
E, seeds of Macuna gigantea, adapted for water dispersal by having an impervious 
coat and contained air-cavity surrounding the embryo (seen in the half-specimen 
on right) ( X 1); F, fruits of Lotus {Nelumbo nucifera), embedded in top of enlarged 
receptacle (both fruits and receptacle are buoyant) ( X £). 


germinated : I am certain of the accuracy of this observation. Again, 
I can show that the carcases of birds, when floating on the sea, 
sometimes escape being immediately devoured : and many kinds of 
seeds in the crops of floating birds long retain their vitality : peas 
and vetches, for instance, are killed by even a few days' immersion 
in sea-water ; but some taken out of the crop of a pigeon, which 
had floated on artificial sea-water for 30 days, to my surprise nearly 
all germinated.' Darwin had already made a conservative estimate 
that the seeds of about one in every ten ' plants of a flora, after having 
been dried, could be floated across a space of sea 900 miles in width, 
and would then germinate '. Although in the light of modern 
knowledge this would seem a rather optimistic guess, at least so far 
as practical opportunities are concerned, there is no reason to doubt 
that odd instances of such off-chance, accidental long-distance 
dispersal do occur from time to time. 

As for the barriers and deterrents to water- and ice-dispersal or 
effective migration, these obviously include any absence of water, 
any obstacle to its movement, or, temporarily, any freezing ' solid ' 
to the bottom. There also seems to be extremely little effective 
interchange between salt and fresh water, while a wide ocean or 
even lake may constitute a barrier to disseminules which cannot float 
and live long enough to cross it ; so may, in addition, a different 
climate which proves unsuitable for the establishment of a trans- 
ported plant. 

Dispersal by Animals (Apart from Man) 

With their obvious mobility and life among plants on which they 
are largely dependent for food and in other ways, many animals are 
important agents of dispersal. Although there are numerous re- 
finements in the method of carriage of the disseminules, there are 
two main_£ategories— those that are c arried externally, by adhesion 
to the surface .ofjt he animal's bod y (the so-called ' ectozoic ' oiv- 
' epizoic ' form of transportation), and those that are carried intern-^-^ 
ally, after^^wallowing (' endozoic ' trans portation) . For this latter 
type of dispersal the seed, fruit, or other disseminule (or container 
of disseminules) is commonly modified by being attractive in appear- 
ance and particularly as food, for example by its bright colour and 
palatable flesh. This should commonly be sweet and juicy when 
ripe, as in Peaches, Figs, Raspberries, and Plums. In addition, 
the embryo or other vital part should be protected from digestion 


by a resistant covering, in which case germination is often hastened 
by passage through an animal. For ectozoic dispersal the dis- 
seminule is commonly adhesive by means of a sticky surface or, 
more often, by its possession of hooks or other devices by which it 
catches on to the fur etc. Anyone who has tried to extract the 
fruits of Burdocks (Arctium spp.) or Beggar-ticks (Bidens spp.) 
from woolly garments will be aware of the effectiveness of such 

Some examples of disseminules modified for dispersal by animals 
are shown in Fig. 27. 

In addition to such ' official ' types of dispersal there is the frequent 
' pecking apart ' by birds : for example, of the seeds or fruitlets 
contained in Apples and Rose-hips. There is also the transport 
of materials for nest-building, and the still more fortuitous adhesion 
of disseminules to the feet, etc., of animals in mud and clay or by 
freezing to their fur or feathers. Thus, for example, Darwin (op. 
^cit., p. 328) mentions removing a considerable amount of clayey 
earth from the feet of Partridges, reporting that in one instance 
around the wounded leg and foot there was ' a ball of hard earth 
adhering . . . weighing six and a half ounces . . . but when . . . 
broken, watered and placed under a bell glass, no less than 82 plants 
sprung from it. . . . With such facts before us, can we doubt that 
the many birds which are annually blown by gales across great 
spaces of ocean, and which annually migrate — for instance, the 
millions of quails across the Mediterranean — must occasionally 
transport a few seeds embedded in dirt adhering to their feet or 
beaks ? ' Quite apart from this, Rabbits, etc., will often drag twigs 
for some distance when these are attached to their fur, and Water- 
fowl have frequently been observed carrying sizeable pieces of 
Pondweeds (Potamogeton spp.) on their backs or around their necks 
— even when in flight. 

(a) Birds. On account of their abundance almost everywhere in 
the world, of the very great distances which many regularly fly, 
and of their consequent power to cross wide expanses of water, 
Birds tend to be the most important. g roup of animal s^rom Jthepoint 
of view of plant di spers al. Although it has been contended by some 
authors that Birds ' fly clean ' on migration, this does not seem to be 
always the case ; indeed, according to Ridley (op. cit. p. 444), it is 
' strongly negatived by much evidence \ Moreover, they are apt to 
' neglect their toilet ' when unwell, and similarly can have materials 
sticking or frozen to their beaks, feet, or feathers when flushed or 




Fig. 27. — Adaptations for dispersal by animals. A, fruits ( X 2t) of Elephantopus 
(left), Cosmos (centre), and Be ggar-ticks (Bidens ^ nffbO, whirh rat-rh on to anima ls; 
B, fruit of Triumfetta "Cx~"i), with hooks causing adhesion; C, sectional view of 
fruit of Peach (Prunus persica), ( X %), showing attractive flesh, protective ' stone ', 
and central seed; D, fruit of Strawberry (Fragaria) (X 1), showing superficial 
resistant ' pips ' enclosing embryos; E, ripe fruit of Nutmeg (Myristica), splitting 
to show seed adorned with attractive coloured ' aril ' ( X £) ; F, fruit of Chinese 
Forget-me-not (Cynoglossum amabile) ( / 4), bearing sticky, hook-like appendages. 


blown out to sea in a gale — or during repeated shorter ' hops ' which 
sooner or later may amount to considerable traverses. Ridley cites 
numerous instances of aquatics, etc., being evidently dispersed to 
isolated ponds and marshes by Water-fowl. Kerner (see the work 
cited at the end of this chapter), like Darwin, secured ' a sufficiently 
striking result ' with fertile seeds in ' the mud obtained from the 
beaks, feet, and feathers of swallows, snipe, ^agtails, and jackdaws 
. . . and when it is remembered that pigeons and cranes traverse 
from 60 to 70 kilometres in an hour, whilst swallows and peregrine 
falcons cover as much as 180 kilometres, it is clear that fruits and 
seeds affixed to these birds may be carried in a very short time over 
several degrees of latitude '. 

An interesting case in point seems to be furnished by the sub- 
antarctic Macquarie Island, situated approximately 650 km. from 
the nearest other land, and supporting thirty-five known species of 
vascular plants. It has recently been contended that all of these 
could well have been, and indeed probably were, brought in by Sea- 
birds since the end of the Pleistocene glaciation. Of these Sea-birds, 
vast numbers inhabit the island and many are known to make long 
flights to South America, New Zealand, and to other subantarctic 
islands. Moreover, many of the habitats on the island, as on 
mountains and in the Arctic, are conveniently ' open ' for the growth 
of immigrants. Unidentified seeds, apparently not belonging to 
any of the local species, have been found on Macquarie Island on 
Black-browed Albatrosses, adhering to the feet and so coated with 
regurgitate that they ' could be carried almost indefinitely in flight 
and could withstand immersion in sea water if the bird alighted to 
rest, yet on landing the seeds would be easily rubbed off ' (Ecology, 
vo1 - 35> P- 57°> October 1954). 

As for endozoic transportation, the effectiveness of this will 
depend not only on resistance to digestion but also on times of 
retention within the Bird's body. Sometimes, especially after 
gorging, seeds may be regurgitated at a distance, without passing 
through the alimentary canal. Kerner found that whereas many 
types of Birds have in their excreta ' under ordinary conditions ' no 
seed capable of germination, some others may void unharmed up 
to 88 per cent, of the small and smooth seeds or fruits eaten, though 
retention in such cases is commonly for only some two or three 
hours. But Ridley quotes a report of Pigeons being shot at Albany, 
N.Y., ' with green rice in their crops, which it is thought must 
have been growing, a very few hours before, at a distance of 700 


or 800 miles ' ; he also gives this distance as the one up to which 
he believes frugivorous Birds have visited very many islands, carrying 
germinable seeds in their viscera and consequently stocking these 
islands with plants. 

Earlier, Darwin had similarly remarked (pp. cit. } pp. 326-7) : 

' after a bird has found and devoured a large supply of food, it is 
positively asserted that all the grains do not pass into the gizzard for 
twelve or even eighteen hours. A bird in this interval might easily 
be blown to the distance of 500 miles, and hawks are known to look 
out for tired birds, and the contents of their torn crops might thus 
readily get scattered. Some hawks and owls bolt their prey whole, 
and, after an interval of from twelve to twenty hours, disgorge pellets, 
which, as I know from experiments made in the Zoological Gardens, 
include seeds capable of germination. Some seeds of the oat, wheat, 
millet, canary, hemp, clover, and beet germinated after having been 
from twelve to twenty-one hours in the stomachs of different birds of 
prey ; and two seeds of beet grew after having been thus retained for 
two days and fourteen hours.' 

What a distance they could have gone in a migrating Peregrine 
Falcon ! 

(b) Mammals. These, among animals, stand next in importance 
to Birds as disseminaters of plants. Except in the case of Fruit-bats, 
which can transport seeds, etc., over stretches of sea much as Birds 
do, their disseminative powers are confined to individual land-masses 
— apart, of course, from traversable shallow or very narrow waters 
or sea-ice in arctic regions (there are no land mammals in Antarctica). 
The Mammals are important dispersal agents of many herbaceous 
plants with small seeds, which they swallow with the foliage, etc., of 
the plants they consume, and are also the main transporters of plants 
with adhesive disseminules. Even though many herbivorous 
Mammals effect such thorough digestion that the vast majority of 
seeds and fruits which they take into their bodies are incapable of 
germination after voiding, there are nevertheless plentiful instances 
of disseminules being excreted unharmed, and we should always 
remember the odd animal that dies suddenly, or is killed and eaten 
by a predator. Ridley (op. ciLjjp. 116) remarks that, in the case of 
stone-fruits, ' almost invariably the seeds pass through the intestines 
of the animal, not only unharmed, but much benefited by the treat- 
ment. Seeds so passed are known to germinate more quickly and 
produce stronger plants than those which have not been swallowed 
by bird or animal and acted on by the gastric or intestinal fluids.' 


Fruits destined to have their contained seeds disseminated by 
Mammals tend to be less conspicuous than those primarily intended 
to attract Birds, e.g. when flying. And even as Birds will devour 
the attractive part of a fruit and scatter the seeds without ingestion, 
so will many Mammals do to large fruits. Arboreal Mammals, such 
as Monkeys, commonly do not eat fruits when they gather them, 
but quietly remove them to a distance — apparently to avoid being 
robbed. If they drop a fruit they do not pick it up, but go on to 
another. Moreover, such types as Squirrels make large winter 
caches that may involve extensive transportation and frequently are 
not eaten in the end. These and many other activities of Mammals 
can help plant dispersal within continental confines. 

Because of their frequently furry coats, Mammals tend to be more 
commonly effective than Birds in the ectozoic transportation of 
adhesive fruits, such as those with hooks or other devices for attach- 
ment. Many fur-coated Mammals wander extensively, or travel far 
on migration, some even crossing wide tracts of sea-ice in the Arctic. 
Apart from being furnished with_obviously e ffec tive hooks or spines, 
some seeds and fruits adhere to animals- by viscid glands, gummy 
exudations, or owing to their wholly sticky nature, while the spikejets 
of many Grasses do so by jagged parts or minutely toothed awns] 
Many other seeds and fruits which are normally wind-dispersed, 
will adhere to animals by entanglement or sticking of their hairs or 
plumes especially when wet. There is, indeed, no lack of means or 
instances of such dispersal, as inspection of one's clothes at the end 
of an autumn walk in a temperate woodland will show. Moreover, 
it should be remembered that animals, like plants, are selective of 
habitat, and tend to keep, as Birds tend to alight, within a single 
habitat-range — so increasing the chances a disseminule would have 
of coming to rest in a place suitable for germination and successful 
establishment,- For instance, the rocky ridges and ravines in boreal 
regions that are inhabited by such birds as Snow Buntings and 
Ptarmigan, which commonly ingest seeds and migrate from one to 
another such area, afford numerous habitats for open-soil Saxifrages 
and Sandworts which may be lacking in intermediate areas. 

(c ) Lower animals. Although most of the Reptiles of the present 
era are carnivorous, some feed on fruits and may disseminate them. 
More important in this respect are freshwater Fishes, many of which 
are vegetable feeders that swallow the seeds of aquatics and semi- 
aquatics, and some of which can migrate overland, usually through 
wet Grass. Of a wide range of seeds or fruits of aquatic plants, such 


as Bog-bean (Menyanthes trifoliata) and Pondweeds, that have been 
fed to Perch and Roach, nearly all germinated after being retained in 
the viscera for one to three days before being passed naturally. These 
fish are liable to be eaten by such predators as Fishing Eagles, 
Herons, and Pelicans, which, after an interval of many hours, either 
reject any contained seeds in pellets or pass them in excreta — often 
still in a viable condition, as was shown by Darwin. The same 
doubtless happens to many Algae, aquatic Fungi, etc. By the time 
such plant material is ejected, the carrying bird may have flown 
many miles. Some of the larger aquatic Crustacea and Mollusca as 
well as, of course, Reptilia, obviously play a part in the dispersal of 
Algae which grow epizoically upon them or their shells ; while on 
land, Snails and Slugs disperse seeds and spores that adhere to their 
bodies or have been swallowed. Indeed, it is said that the spores 
of some Fungi will only germinate after passing through a Slug ; and 
when the latter is eaten by a Toad, Bird, or other predator, the 
possibility occurs of far more extensive dispersal. 

Insects are probably the most important of the groups of lower 
animals in the matter of plant dispersal, especially of very small 
bodies such as fungal spores. Transport is commonly by swallowing 
and ' passing ' in the excreta, by carrying to their nests for food, and 
by adhesion. Locusts are said to afford examples of the first method, 
sometimes over considerable distances, and ants frequently transport 
seeds with edible appendages, while flies and many other insects 
often carry spores of cryptogams adhering to their bodies — especially 
when the latter are densely hairy. Further instances are the well- 
known transmission of Fungi- and Bacteria-engendered diseases by 
insects, as well as important viruses (such as those of Potatoes) 
having aphid vectors. 

It seems desirable here to treat briefly the subject of pollen. 
As we saw in Chapter II, pollen is composed of vast numbers of 
microscopic ' grains '. These, though capable of producing on 
germination only a tiny particle of plant, and hence scarcely to be 
considered as true disseminules, nevertheless carry the potential 
male gametes and, in them, the genes introducing hereditary 
characters. As it is now known that transport of pollen can in some 
circumstances take place naturally over many hundreds of miles, 
and that given suitable conditions some detached pollens can live 
for many months, it seems conceivable that by this means heritable 
characters may be transported vast distances. To be sure, the 
grain has to find its way to a receptive female stigma to have any 


chance of effective survival. But when it' is recalled that pollen 
grains are formed each year in trillions of trillions, and that a single 
pollen ' bullet ' finding its stigmatic billet in a millennium might 
suffice to carry thither the genes of any subspecific characters it may 
possess, the possibility can scarcely be denied of what has facetiously 
been termed ' absent-treatment hybridization '. Wind and insects 
are the chief transmitting agents of pollen, though carriage is also 
effected by some other animals (especially small Birds) and by 
water. Most of the strikingly beautiful features of flowers, as well 
as their possession of nectar and scent, are adaptations to attract 
insects to gather pollen for the purpose of cross-fertilization, and so 
it is to be expected that this is very commonly effected, though 
chiefly over rather short distances. 

Dispersal by Human Agency 

There can scarcely be any question that Man is the most active 
agent of vegetational change — including plant dispersal — of modern 
times. He is the greatest despoiler of forests and causer of erosion, 
dispersing weeds as well as growing crops. As he travels about the 
world in greater and greater numbers and with ever-increasing speed 
and ease, he is always transporting the disseminules (or sometimes 
transplanting whole individuals or groups) of plants either intention- 
ally or unwittingly. Also of vast importance is his indirect effect, 
through the pasturing of his domestic animals or his disturbance 
of natural communities of herbivorous animals. As a result, there 
are few parts of the world where the vegetation and its component 
flora do not bear the stamp of Man's interference, and quite a few 
areas, for example in Hawaii and Ceylon, where the native plants 
have been largely ousted by alien ones. In general, however, unless 
there is some drastic disturbance of the natural vegetation, recently 
introduced plants fail to compete successfully with the native 
dominants and, consequently, take only a minor part in the con- 
stitution of most plant communities. Often these aliens are limited 
almost entirely to burned-over or otherwise cleared areas — such as 
waysides and abandoned fields, which characteristically support 
hosts of weeds. 

Between the extremes of those plants, such as many horticultural 
strains, which are restricted to gardens and need constant tending, 
and those which, following introduction, have so thoroughly estab- 
lished themselves by natural agency that they are distinguishable 


from the indigenous flora only by their known history, we see all 
manner of degrees of success in establishment. Some aliens flourish 
for a time and then disappear, while others, after many years of 
restriction to one locality, suddenly burst forth all over a countryside. 
A notable example of the latter category is the Oxford Ragwort 
(Senecio squalidus), which was introduced into the Oxford Botanic 
Garden late in the seventeenth century but scarcely spread at all 
until late in the nineteenth century, when it started migrating along 
the railways. Thereafter, migration proceeded so extensively that by 
the nineteen-twenties it became known from the vicinity of railways 
in other counties, and when, in the nineteen-thirties and -forties, the 
present writer was in charge of the botanical collections at Oxford, 
it was apt to be sent in from quite remote districts of Great Britain 
as a curiosity or for identification. 

It will be sufficient — without going into detailed examples which 
could fill whole chapters — to indicate here^ome ofihejmain methods 
by which Man introduces plan ts to new areas and, often^new 
countries and even continents_(for it is said that the majority of 
alien plants in Australia and New Zealand come from Europe). 
In addition to intentional transport of desirable plants for agri- 
cultural, horticultural, forestral, medicinal, or other purposes, weeds 
are often dispersed unwittingly with the seeds of vegetables, cereals, 
and garden flowers^ as well as with pot plants and in making trans- 
plants. All manner of disseminules and whole plants are dispersed 
accidentally (but quite commonly) by land or water traffic, garbage 
removal, and in baggage and soil transportation, while admixture 
in animal fodder, litter, and manure are other extensive means of 
transport. Dispersal used to be widely effected in ships' ballast and 
still is in many packaging materials, often to the far corners of the 
earth, as it is also in bird-seed and building material, or as algal 
growth attached to ships' hulls. Other sources of dispersed dis- 
seminules are timber and drug and spice materials — such as Caraway 
seeds imported by the Danes to Greenland for flavouring bread, with 
the result that the plant, Carum carvi, is now common around many 
of the settlements. Indeed, very many kinds of commercial export- 
import traffic must involve the carriage of disseminules, some of 
which evidently lead to fresh introductions ; the same is true of 
personal travel, for people often carry a considerable range of seeds 
and fruits about their clothing, and, doubtless, greater numbers of 
microscopic spores. 

In this connection air travel may be particularly effective, for one 


steps on to an aircraft in one continent and off it in another, often 
with little movement meanwhile to brush off adhering seeds and 
fruits. Much the same may be true of transported animals, which 
are always apt to carry seeds and fruits in their wool and fur, or 
otherwise about or in their bodies, and evidently account for many 
plant introductions. Moreover, there is practically no limit to the 
number and diversity of seeds and fruits that adhere to men and 
women when they fall or merely walk in mud and clay, to be brushed 
or picked off later, or are transported by them for food or as curios — 
often to be discarded at a distance. And to the abilities of some 
seeds to pass through the human digestive tract unharmed, the 
' spontaneous ' growth of Tomato plants in sewage farms bears 
ample testimony. Finally, with their hosts are frequently carried 
parasitic (and also saprophytic) species. That such spread of plant 
diseases can be very serious is indicated by the rigid restrictive 
measures adopted by many governments against the importation of 
living plants. 

The great differences often observed in the actual migration of 
thus ' artificially ' introduced plants are, however, probably due 
more to the adaptability of the species to local environments than 
to the dissemination itself, essential though this is. As we shall see 
in Chapter VI and elsewhere, plants tend to be adaptable when thev 
are variable — in habitat requirements as well as in form. Accord- 
ingly, some familiar European weeds, such as Shepherd's-purse 
(Capsella bursa-pastoris), Common Chickweed (Stellaria media agg.), 
and the little grass Poa annua, have become practically world-wide 
without having any special adaptations for long-distance dispersal, 
whereas other plants the disseminules of which are doubtless more 
commonly, and sometimes more widely, carried, are still relativelv 
restricted in their geographical area. Often the climatic, soil, or 
other local conditions are unsuitable ; or the competition of native 
plants is so severe that ' open ' habitats have to be found — especially 
by weeds. Such open habitats are commonly due to Man's acti- 
vities, soon becoming closed over with vegetation when abandoned, 
so that the colonizing aliens become restricted or often ousted. In 
time the signs of human interference may virtually disappear, 
though, as has already been emphasized, such interference is nowa- 
days so widespread and drastic in various ways as to constitute the 
most active agent of vegetational change in the world. Moreover, 
as compared with earlier times, the barriers to dispersal by human 
agency are greatly diminished now that men can (and frequently do) 


travel to almost all parts of the world in a matter of days, and 
traverse vast distances in a very few hours. 

Mechanical Dispersal 

Although it is usually effective over only short distances, mech- 
anical propulsion or even extensive growth can be of distinct 
advantage in migration. Thus plants which shoot out their dis- 
seminules can thereby launch them into a goodly wind or on to a 
passing animal that will carry them for miles. And often it is 
agitation by wind or an animal which sets off the explosive mech- 
anism. Furthermore, the aggressive growth of overground runners 
(Fig. 28, B) and underground stems (rhizomes, see Fig. 28, A) often 
give plants a distinct advantage in competition over their neigh- 
bours, so that when, as is often the case, the peripheral growth is 
detached as a separate plant, for example by the death of the parent, 
it may be established at an appreciable distance ; and such distances 
mount up usefully through the generations. As examples, Ground- 
ivy (Glechoma hederacea) can trail a distance of 20 feet (about 
6 metres) along the ground, and Elms can reproduce by suckers from 
underground roots at fully 50 yards (about 46 metres) from the 
parent tree. Even such growth as that of the Walking Fern shown in 
Fig. 22 leads, in due course, to a worthwhile amount of dispersal. 

Particularly effective are the explosive spore-discharging mech- 
anisms of some Fungi, which, usually on sudden rupture to relieve 
stresses, may shoot their spores or spore-producing organs in some 
instances as much as 15 feet. However, in the case of spores, a 
tiny distance to take them into the free air is often sufficient to 
launch them in atmospheric currents that may carry them practically 
anywhere. Also capable of being shot out for distances of as much 
as 3 feet are the bulbils of some Club-mosses (Lycopodium spp.). 
Better known, however, are the explosive mechanisms of some fruits, 
of which examples are shown in Fig. 28 and more may be cited. 
The ' records ' seem to be held by species of a genus of small 
parasitic Mistletoes (Arceuthobium), followed by tropical American 
trees of the Spurge family (Euphorbiaceae), particularly the Sand-box 
Tree (Hura crepitans), which can throw its seeds more than forty 
feet, and Para Rubber (Hevea brasiliensis), whose performance is 
nearly as good. The explosion of a Hura fruit is spoken of as a 
1 regular detonation '. In a similar manner, on drying of the fruits, 
even small herbaceous members of the Spurge family often shoot 
out their seeds for a distance of a dozen or more feet. In the case of 




Fig. 28. — Dispersal by extension of growth or by mechanical propulsion, etc. 
A, horizontal rhizome of a Grass (the sand-binding Marram Grass, Ammophila 
arenaria (x about \)\ B, Strawberry (Fragaria) runner establishing daughter 
plant ( x ?) ; C, fruit of a Balsam {Impatiens balsamina), which explodes and 
scatters the seeds (x 1); D, seed dispersal in the Squirting Cucumber {Ecballium 
elaterium {< about |); E, ripe fruit of Pansy {Viola sp.), showing seeds ready 
to be, and being, shot out (x 2); F, over-ripe fruit of Geranium {Geranium 
sp.), showing slings which have thrown out the seeds ( X about 2^). 

one species of Arceathobium the tiny bullet-shaped seeds have been 
reported to travel over 66 feet (about 20 metres) from a point 8 feet 
above the ground, and in one instance large numbers were collected 
from the roof of a cabin one-quarter of a mile (about 402 metres) 
from the point of liberation — presumably after transportation by 
wind, though the seeds are also viscous and apt to be carried by birds. 
Such a combination of explosion and adhesion is utilized also by 
the Squirting Cucumber {Ecballium elaterium), shown in Fig. 28, D. 
When the fruit is ripe, it breaks from the stalk, and through the hole 
thus left the internal pressure is relieved by the seeds being ejected 
with an abundance of mucilage and with such force that they com- 
monly fly for several feet through the air. A slight touch will send 
off the ripe fruit, so that a passing animal is liable to receive a broad- 
side — and, incidentally, carry the adhering seeds much farther. 
There are many varieties of this type of turgor-engendered explosion, 
another being exhibited by certain Cresses (Cardamine spp.) that 
have explosive pods of which the narrow valves, on being touched 
when ripe, suddenly curl outwards with some violence, shooting out 
the seeds — sometimes for more than 2 feet. In the case of such 
diminutive plants, this is ample to take the seeds away from the 
parental sphere of influence. Notable among the rather many and 
diverse plants which do this sort of thing are the Balsams (Impatiens 
spp.), in which the wall of the fruit is made of three layers of which 


the innermost consists of large turgid cells. ' When ripe, especially 
if touched, the wall of the fruit suddenly separates into five segments 
that curl inwards violently (cf. Fig. 28, C), shooting out the seeds 
— in some species for fully 20 feet. In the Wood-sorrels (Oxalis 
spp.) the ripe fruit suddenly splits lengthwise or by lateral slits 
when touched, shooting out the mucilage-covered seeds. 

In many fruits the explosion that leads to a forcible ejection of 
seeds is caused by stresses set up on drying. The audible cracking 
of the pods of some members of the Pea family (Leguminosae) is of 
this nature, the two valves of the pods (for example, when drying 
in the sun) suddenly separating with often a violent spiral twisting, 
and forcibly ejecting the hard and smooth seeds. Familiar European 
examples of this are furnished by the Gorses (Ulex spp.). The 
action is due to a hard layer of strongly thickened, elongate cells 
lying transversely, and to which the softer tissues offer little resist- 
ance. The distances to which seeds are shot by these means vary 
greatly, but in some instances are said to exceed 40 feet and at 
least rival the ejections of Hura and Hevea. Thus the turgor- and 
drying-induced methods may be about equally effective. 

Many species of the familiar genus Viola, including some wild 
Violets, in which the fruit splits into three boat-shaped valves 
(Fig. 28, E), shootout their seeds for up to 15 feet as a result of unequal 
drying of the layers of the fruit wall. This drying causes a curving 
of the sides of the valves and the consequent pressing of the glossy 
seeds together — until they ' pip out ', one after another, being often 
further dispersed by rainwash. Also dispersed on explosion of the 
hard ripe fruit are the seeds of almost all members of the family 
Acanthaceae — sometimes to nearly 30 feet from the parent — and 
those of Claytonta, Montia, some Phloxes, and the Witch-hazels 
(Hamamelis spp.). In the last instance the drying fruits may exert 
such pressure on the seeds that these are discharged, like miniature 
bullets, to distances of up to 40 feet — again rivalling Hura and Hevea. 

In many members of the Geranium and Stork's-bill family 
(Geraniaceae) the fruit suddenly splits into strips which curl up and 
act as slings (Fig. 28, F) to throw out the seeds, which may travel 
as much as 20 feet. Some fruits are also effectively dispersed by 
mechanical propulsion, including those of Flat-figs (Dorstenia spp.) 
which are embedded in the large fleshy receptacle that shrinks on 
drying, setting up pressures which lead to the tiny fruits being 
forcibly ejected. The spores of many Ferns are well known to 
be discharged forcibly into the air by the springing backwards of 


part of the capsule when it has dehisced and attained a certain degree 
of desiccation, while the movements of teeth of Mosses are hygro- 
scopic, curving backwards to open the capsule and disseminate the 
spores. These actions take place chiefly in dry weather when 
conditions are best for dispersal. Also able to move as a result of 
hygroscopic changes are many awned fruits, etc. Finally, it should 
be recalled that the spores or even the whole bodies of many of the 
lower cryptogams are actively mobile, swimming by means of flagella 
being particularly common among them. 


When we reflect that in many species of flowering plants, such as 
Flixweed {Sisymbrium sophia) and Pigweed (Amaranthus retroflexus). 1 
a single individual may produce a million or more seeds in one 
summer, and that some cryptogams, such as the Giant PufTball 
(Ly coper don (Cahatia) giganteum), may produce as many as several 
million million spores, and yet none overruns the world, 2 it is obvious 
that only an infinitesimally small proportion of the plant disseminules 
produced ever attain their real biological raison d'etre. To realize 
its full potentiality, a propagule must develop into an adult which 
in turn reproduces. This stupendous mortality is due to the action 
of various types of barriers — either to dispersal or to actual survival 
— of some of which we have already seen examples as applied to 
particular agents of dispersal. They are of four main types : 

(1) Physiographic, due to features of the earth's surface. The 
most obvious of these for terrestrial plants are expanses of water, and, 
for aquatic plants, bodies of land. Another physiographic barrier 
is afforded by mountains — both directly by constituting a mechanical 
impediment, and indirectly by changing climatic and allied conditions 
such as air temperatures and currents. Many local winds are 
caused by a combination of physiographic and climatic factors, and 
constitute virtual barriers to dispersal in one direction even as they 
may aid it in another. 

1 An individual of this annual species has been known to produce an estimated 
2,350,000 seeds. 

2 It is said that an individual Giant PufTball can produce 7,000,000,000,000 
spores, and it was calculated by the late Professor A. H. R , Buller (Researches 
on Fungi, vol. Ill, 1924) 'that, if every spore of this puff-ball had germinated 
and given rise to a puff-ball like its parent, and if every spore of the second- 
generation puff-balls had likewise germinated and given rise to a puff-ball like its 
parent, then, at the end of these two filial generations only, there would have 
come into existence a mass of puff-ball matter equal to 800 globes the size of 
the planet on which we live ! ' 


(2) Climatic, involving different temperature, humidity, light, and 
other conditions. Owing to the close dependence of plants on 
climatic conditions, zones of vegetation and climate tend to cor- 
respond with one another, the climate commonly determining the 
general limits of a plant's distribution. A change of climate, such 
as a migrating plant is apt to find in a new land, often constitutes a 
very real or insuperable barrier — not only as a whole but, very 
often, as to one or other of the climate's component factors, which 
may react in a particular way on the plant's physiological make-up or 
a vital part thereof. Moreover any condition, including lapse of time, 
which proves lethal to disseminules may constitute a major barrier, 
lack of viability being an important factor militating against migration. 

(3) Edaphic, due to features of the soil. These are again various, 
involving physical structure, chemical composition, moisture con- 
tent, temperature conditions, or even content of living organisms, 
any one of which alone can prevent a disseminule from establishing 
a plant in a new area, even if it germinates quite successfully. 
Either in combination or separately, edaphic conditions tend to 
limit the distribution of plants (and, of course, vegetation) rather 
drastically within the main climatic belts — commonly to particular 
habitats, which may be narrowly prescribed in their type and of 
very limited extent. Absence of the suitable habitat, or at least of 
the required conditions, is apt to constitute an insuperable barrier 
to successful migration. 

(4) Biotic, due to living organisms, including other plants. The 
competition for space, light, water, etc., of other plants already 
established in an area and growing in reasonable equilibrium with 
local conditions, is apt to constitute an insuperable barrier to the 
successful establishment of newcomers, as is grazing or other dis- 
turbance by animals (including Man). Consequently widespread 
immigration is largely limited to more or less ' open ' areas, such as 
cliffs, sands, and disturbed ground — where other sets of barriers 
come into play. If they did not, virtually every scrap of soil or ray 
of light would probably be utilized ; for the struggle for existence 
is a very real and desperate one, taking place chiefly between organ- 
isms where the general conditions for life are good, and predominantly 
with the physical factors of the environment where conditions are bad. 

The magnificent perseverence and virtual ubiquity of plant life 
are vividly exemplified to the author by the myriad Diatoms which 
may so impressively if variously colour ice-floes on arctic seas. 
During long flights over the North Polar Basin and elsewhere he 


has observed such ' dirty ' floes in many places up to almost the 
highest latitudes (the ice immediately around the North Pole appears 
to be ' clean '). From the air, some of the browns and yellows of 
these floes seem not far removed in colour-effect from some barren 
arctic limestones — as the author and his pilot had occasion to remark 
once in 1946 when flying over Foxe Basin and sighting an unexpected 
island of limestone which turned out to be some 90 miles long and 
nearly as wide. It was officially ' discovered ' two years later by 
the Royal Canadian Air Force and added to the world's map as 
' Prince Charles Island ', being, with its neighbours which were also 
noted on that 1946 occasion, evidently the last major land discovery 
or confirmation to be made in the world. Subsequent exploration 
failed to reveal any unexpected features of plant life. Until the 
advent of general air travel not so long ago, many areas even in 
comparatively low latitudes were difficult or at least tedious to visit. 
But now in superb antithesis we are looking to other planets for 
explorational opportunities, and the prospects of space travel are 
advancing so rapidly that the author is prompted to guard himself 
by remarking that the present volume is concerned purely with 
vital phenomena as we know them on the Earth and in its immediately 
surrounding atmosphere — regardless of any possibilities elsewhere. 

Further Consideration 

H. N. Ridley. The Dispersal of Plants throughout the World (Reeve, 
Ashford, Kent, pp. xx -f 744, 1930). A monument of usually 
authoritative information on the methods and effectiveness of plant 
dispersal, covering almost all aspects of the subject. Its author died 
recently at well over 100 years of age, maintaining that such studies, 
of which there are still not nearly enough, are fine preservers of life. 
harles Darwin. The Origin of Species by means of Natural Selection, 
sixth edition, with additions and corrections (Murray, London, 
pp. xxi + 458, 1873). 

H. B. Guppy. Plants, Seeds, and Currents in the West Indies and Azores 
(Williams & Norgate, London, pp. xi + 531, 1917). A work of 
wider implication than its title suggests. 

A. Kerner (von Marilaun), & F. W. Oliver. The Natural History of 
Plants, vol. II, pp. 790 et seq. (Blackie, London, 1895). 

Sir E. J. Salisbury. The Reproductive Capacity of Plants (Bell, London, 
pp. xi + 244, 1942). £ 

L. V. Barton. Seed Preservation and Longevity (Leonard Hill, London 
— in press). 

C. T. Ingold. Dispersal in Fungi (Clarendon Press, Oxford, pp. 
viii + 197, 1953). 

Chapter V 


In Chapter II we gave a brief general account of each of the main 
classes of the plant kingdom now living, mentioning that the sequence 
used was probably indicative of evolutionary history at least in broad 
outline. The classes treated were usually those most important as 
components of vegetation at the present time, regardless of their 
significance in the past. Those omitted even included some that 
had been vastly important in earlier ages but had subsequently 
become extinct or nearly so. 

Evolution is a continuous process, that started with the earliest 
forms of life and still goes on abundantly today. And even as the 
numerical representation and importance of a particular group of 
organisms may go down as well as up, so may evolution manifest 
itself in simplification or decline as well as advance. This is parti- 
cularly evident in many parasitic forms of both plants and animals. 
Nevertheless the general trend is towards advancement in complexity 
if not always in size, economy in the use of material being also 
important ; with these tendencies and the need for adaptation to the 
environment and other circumstances constantly in mind, we can 
place the forms known to us in a sequence that seems most likely 
to be the actual one of their own evolution. This will be done 
briefly below for those groups that are important as fossils, regard- 
less of the significance of relatives living today, but with close 
reference to these in order to link relationships in the mind's eye. 

It is necessary here to give an outline account of the geological 
time-sequence. In this, four major eras are defined, the last three 
of which are divided into several periods or epochs each. It is now 
believed that the earth had its beginnings about 4,500,000,000 years 
ago. The latest measurements indicate the age of the oldest known 
rocks to be about 3,300,000,000 years, and the era from that time 
to about 550,000,000 years ago constitutes the pre-Cambrian. This 
era has often been divided into two — the ' older ' Archaeozoic (of 
metamorphic and igneous rocks) and the ' younger ' Proterozoic (of 



sedimentary rocks). At least in the latter time, relatively simple 
Algae and Sponges and apparently also Fungi and Bacteria were 
widespread. The pre-Cambrian was followed by the Palaeozoic era, 
of invertebrates and Fishes and large Pteridophytes. It extended 
for about 360,000,000 years from the Cambrian period through the 
Ordovician, Silurian, Devonian, and Carboniferous (Mississippian 
and Pennsylvanian) periods to the Permian period, which ended 
about 180,000,000 years ago. Next came the Mesozoic era, which 
extended for some 130,000,000 years through the Triassic, Jurassic, 
and Cretaceous periods and was the great era of Reptiles and Gymno- 
sperms. Finally, extending over the last 60,000,000 or so years, has 
been the Cainozoic (Cenozoic) era, of Angiosperms and Mammals. 
This is commonly considered to consist of two periods, the Tertiary 
(made up of the Paleocene, Eocene, Oligocene, Miocene, and Pliocene 
epochs) and the Quaternary. In the higher latitudes and altitudes 
this last period consisted of alternating glacial and interglacial times 
and may be referred to as the Pleistocene epoch, the ' recent ' being 
the time since the last ice recession took place, although some con- 
sider this a mere interglacial. The Quaternary period has extended 
over perhaps the last 1,000,000 or so years 1 and has seen the advent 
and ascendancy of Man. A chart showing the eras and main periods 
etc. is given in Fig. 37 (p. 145). 

Groups of Fossil Lower Plants 

We can only guess at the form of the first living organisms, which 
were probably not distinct as either plants or animals, but must have 
possessed the powers of deriving energy from outside sources and of 
sustaining themselves. Presumably they were microscopic bits of 
naked protoplasm far simpler than any organisms of which fossils 
are known. From such a source sprang the ' tree of life ', near the 
bottom of which the Bacteria appear to remain. These organisms 
play such essential roles as agents of decomposition that it is difficult 
to conceive of the balance of nature being maintained without them, 
and indeed there is evidence that they were in existence in very 

1 According to F. E. Zeuner's Dating the Past : an Introduction to Geochronology, 
third edition (Methuen, London, pp. xx + 495 and 24 additional plates, 1952). 
Other modern estimates range from one-half to double this total, a difficulty being 
to decide at what point in time the Quaternary began. Very recently, Professor 
Zeuner has suggested (in litt. 1957) that ' an estimate of 600,000 years for the 
period from the First [Pleistocene] Glaciation onwards is a reasonable one '. 




early geological times. Thus in rocks far back in the pre-Cam- 
brian 1 are found supposed signs of Bacteria in the form of filamentous 
chains and minute spherical bodies, often associated with slender 
branching filaments and other remains believed to be of Blue- 
green Algae (cf. Fig. 29, C). What appear to be the oldest known 
structurally preserved organisms that clearly exhibit cellular differ- 
entiation and original carbon complexes are in pre-Cambrian sedi- 
ments of southern Ontario and represent Blue-green Algae and simple 
forms of Fungi or possibly Algae (see Fig. 29, D). It seems probable 
that their age exceeds 1,000,000,000 years, and it may quite likely 
be nearer to 2,000,000,000 years. From later on, and at least 
beginning with the Devonian around the middle of the Palaeozoic 
era, 1 there are plentiful indications that Bacteria were practically 
ubiquitous, as indeed they are today. Nor is there evidence of major 
evolutionary change in their structure through those many millions 
of years. 

Algae, which among living plants probably rank next in importance 
to vascular ones in the formation of contemporary vegetation, are 
also significant for the roles they appear to have played in the 
formation of petroleum and limestone. As a result of Algae obtain- 
ing carbon dioxide for photosynthesis from (soluble) calcium 
bicarbonate, the relatively insoluble calcium carbonate was left as a 

1 See previous page and Fig. 37 concerning the geological eras, etc., with their 
dominant forms of life and supposed ages. Fig. 37 also indicates the approximate 
relative development of the main plant groups at different times. It is now thought 
that the earth may be of the order of 4,500,000,000 years old, and that life may 
have begun on it about half-way along the time to the present. 






Fig. 29. — Some primitive plant fossils. A, Collenia undosa, a Proterozoic fossil 
believed to have been formed by the action of one or more Blue-green Algae 
(after Walcott) (/ |); B, Newlandia concentrica, a Proterozoic fossil apparently 
formed by the action of a Blue-green Alga (after Walcott) ( X f ) ; C, a Proterozoic 
fossil, apparently a colonial Blue-green Alga consisting of aggregations of filaments 
in globose sheaths (courtesy of E. S. Barghoorn and Science) (x about 200); D, 
a Proterozoic fossil of fungal or possibly algal type, showing spores and non- 
septate hyphae (courtesy of E. S. Barghoorn and Science) (x about 400); E, a 
Cambrian Alga, Dalyia racemata (after Walcott) (x about 3); F, a Cambrian 
Alga, Marpolia aequalis (after Walcott) (X about 2). 


residue and became deposited as limestone.' Limestones believed 
to have been formed in this manner by Algae occur in extremely 
early rock formations, and apparently such deposition of calcium 
carbonate as a result of algal activity has been going on ever since. 
However, owing to such features as their generally soft bodies, Algae 
often leave no traces of their original cellular structure, and so as 
fossils they are difficult to recognize with certainty. In any case 
there seems no doubt that, well before the end of the Cambrian 
period, not only an abundance of Blue-green but also large numbers 
of Green and some Red and probably Brown Algae had evolved. 
Certainly all of the first three groups were plentiful in the Ordovician. 
Two examples of fossil Algae of the Cambrian period are shown in 
Fig. 29, E and F. Although Desmids and even Stoneworts are now 
known to go back into the Palaeozoic, being present in the Devonian 
period, the Diatoms appear to be of more recent origin, none being 
known before the Jurassic period in the middle of the Mesozoic era. 
It may well be that many groups of Algae have remained evolution- 
ary static throughout the time since their remote ancestors laid 
down some of the earliest fossils of which w r e have knowledge. 

It was indicated above that the Fungi as a group are also very old, 
and there is no reason to doubt that they have acted as scavengers 
throughout their long geological past, even as they act today. Indeed 
it appears that fossils were chiefly formed when deposition of plant 
material took place under conditions unfavourable to fungal growth, 
so that the usual destructive activity of Fungi was evaded. Well- 
preserved fungal mycelia and spores have been found in the tissues 
of vascular plants as far back as the Devonian, and Fungi apparently 
occur in sedimentary deposits of much earlier date (see p. 130). 
Such discoveries have not, however, shed any clear light on the 
origin of the Fungi, which have long been believed to have evolved 
from Algae through the loss of chlorophyll. However, some authori- 
ties now hold that Fungi were derived from a distinct group of 
primitive organisms, the similarities with Algae being due to parallel 
evolution, and it may well be that, unlike the various members of 
most other groups, Fungi had no common starting-point but origi- 
nated at different times and in various groups. Fossils of higher 
Fungi do not appear with certainty until the Cretaceous, and no 
indubitable early fossils of Lichens are known. 

The Nematophytales comprise an extinct group of spore-producing 
Silurian and Devonian plants of uncertain relationship, in which the 
plant body was composed of a system of interlacing tubes. It may 


be that their affinity is with the Algae, and it even seems conceivable 
that they may represent the long-sought link between that group 
and the lowest land-plants. A further link may be forged by 
certain cellular forms producing firm- walled spores, which forms, 
like some Nematophytales, have a covering cuticle resembling that of 
land- plants, but which seem to belong to a relatively advanced group 
without actually being typical land-plants in other respects. 

Fossil remains of Bryophyta are not common, probably owing to 
the fragile nature of the plant body. Some Liverworts are known 
from as earlv as the Carboniferous towards the close of the Palaeozoic 
era ; traces of Mosses have also been found in rocks laid down before 
the end of the Carboniferous. From the Triassic period onwards, 
fossil Bryophytes tend to become less rare, so that a considerable 
number are known from the Pleistocene. It now seems that fossil 
Bryophytes throw little if any light on the problem of the origin of 
vascular plants, and that there is no justification for thinking them 
ever to have served as intermediate stages in the evolution of 
higher plants. Instead, these presumably evolved from Algae, as 
also with little doubt did the Bryophytes — but in this latter instance 
without going much ahead. Consequently they are little changed 
to this day, although they are now numerous and often ecologically 
successful within the limits prescribed by their relative diminutive- 

By the close of the Silurian period there were undoubted land- 
plants, the primitive aquatic or semi-aquatic Algae (perhaps through 
more advanced types such as the Nematophytales or others of which 
we have no knowledge) having apparently come out on land and 
given rise to vascular forms. As we saw in Chapter II, these last 
are characterized by the possession of a conducting system com- 
posed essentially of wood and bast elements ; nor is it by any means 
impossible that such a system was developed among marine Thallo- 
phytes, the more adaptable of which may gradually have become 
transformed to withstand permanent life on land. At the same time 
their holdfasts could have developed into rhizome-like structures 
bearing rhizoids. For such are the earliest known land-plants, and 
so may the great ' subaerial transmigration ' have taken place. 

The earliest known plants that were clearly adapted to life on land 
belong to the class Psilophytineae, which was probably more primitive 
than any of the other Pteridophytes, and of which some reconstructed 
examples are shown in Fig. 30, A and B. They range from the 
middle Silurian to the upper Devonian periods of the Palaeozoic era, 

Fig. 30. — Some Psilophytineae. A, two species of Rhynia, showing the sporangia 
at the ends of the hranches (after Kidston & Lang) ( X $•); B, Psilophyton princeps, 
showing rhizomes below and, above, young branches uncurling at left and branches 
bearing sporangia at right (after Dawson) (X probably about To); C, Psilotum 
triquetrum, showing the habit (x I) and, enlarged, on left, part of a branch with 
sporangia. A and B are reconstructed fossils, C is drawn from life. 



and appear to be represented by a very few allies living today. Of 
these the best known is Psilotum (Fig. 30, C), which is widespread 
in warm regions. Both extinct and living members are branched, 
with naked or spiny stems having cuticle and stomata on the surface, 
and sometimes small simple leaves. They form spores of one size, 
produced in characteristic sporangia. Further details regarding the 
Psilophytineae and also some other groups of ancient vascular plants 
of often obscure relationship, may be obtained from such works on 
fossil botany as are cited at the end of this chapter, though com- 
parisons will show how difficult it sometimes is for authorities to 
agree. Thus the present group are sometimes given the status of 
a division, as the Psilophyta (or Psilopsida). 

The Equisetineae (Horsetails) have also a very ancient history, the 
present-day representatives, already dealt with in Chapter II, being 
mere depauperated relics of a once large and important group that 
flourished at least as far back as the Devonian. With their extinct 
fossil representatives they are sometimes given the rank of a division, 
under the name of Arthrophyta (or Sphenopsida). They show 
plentiful adventitious roots and small whorled leaves, and sometimes 
secondary thickening. Of this major group there are five subsidiary 
groups or orders, of which the most important are : (1) the lowly 
Sphenophyllales, which had slender reclining stems and expanded, 
wedge-shaped or lacerate leaves usually less than 2 cm. in length ; 
(2) the Calamitales, which were like giant Horsetails, attaining heights 
of some tens of feet and with their jointed, hollow stems sometimes 
exceeding 20 cm. in diameter ; and (3) the Equisetales, which were 
characterized by their slender, jointed stems and were altogether very 
like the representatives living today. Although the Equisetales 
appeared only in the Carboniferous, they were and are closely allied 
to the earlier Calamitales, and consequently Equisetum may be 
regarded as the oldest living type of vascular plant ; indeed some 
authorities maintain the Calamites and Equiseta in the same order. 
Probably both groups arose from some common source, the 
Equisetales lingering on to the present day without major changes 
and representing the end of a once virile line whose ecological 
aggressiveness still saves it from extinction. It should be noted 
that in some Calamitales there was differentiation into large mega- 
spores and small microspores. A living Equisetum is shown in 
Fig. 15, and in Fig. 31 may be seen fossils of typical members of 
the other two main groups. 

Living representatives of the Lycopodineae (Club-mosses and 



their allies) are described in Chapter II arid illustrated in Fig. 16. 
They, too, have a long fossil history, both as the small herbaceous 
types which we know today and as huge trees that flourished chiefly 
in the Carboniferous period, when they evidently composed a large 
part of the vegetation, at least in coal-forming swamps. Subse- 
quently, near the beginning of the Mesozoic era, the tree types 
appear to have become extinct, much as in the case of the giant 


Fig. 31. — Fossil Equisctineae. A, a Calamite, Catamites suckowi (x about l)\ 
B, a Sphenophyll, Sphenophyllum emarginatum ( < about 1). (Both after Zeiller.) 


Horsetails. Outstanding examples are Lepidodendron and Sigillaria, 
shown in Fig. 32, which were characterized by extensive rooting 
systems of a unique kind, spores of two sizes, and usually tall and 
straight, woody trunks covered with the scars left by the spirally- 
arranged leaves. These last were narrow and grass-like but ligulate, 


,:;>;|4 ' P : ' " 

» 1 . (ft . : .,- ., 

W ;i , 

Fig. 32. — Fossil Lycopodineae. A, Lepidodendron, two figures on the left, and 
Sigillaria, five figures on the right, showing also the characteristic rooting system 
(after Grand'Eury) (the tallest is about 16 metres in height); B, Lepidodendron 
lycopodioides, showing the small leaves and leaf scars (after Zeiller) ( X about §). 

and in Sigillaria are said occasionally to have exceeded 50 cm. in 
length. Some of the earlier groups quite likely go back to Silurian 
times, and probably evolved from the psilophytinean complex ; later 
ones became highly specialized trees, sometimes with seed-like 


organs. But, having become structurally modified for life in swamps, 
with extensive tissues for gaseous exchange and very little for water 
conduction, they were apparently unable to adapt themselves to the 
more adverse climates following the Carboniferous. Meanwhile 
the small herbaceous types survived and gave rise to those persisting 
at the present time, even though they apparently represent another 
evolutionary end-line. Like the Psilophytineae and Equisetineae, 
this group is sometimes given the rank of a division, as the Lycopsida ; 
for it is now realized that these groups of so-called ' fern-allies ' 
represent separate lines of development as far back as it is possible 
to trace them. 

Of Filicineae, the Ferns, characterized by large complicated leaves 
and gaps in the vascular cylinder, there are plentiful fossils — which 
go well back into the Devonian in the case of the long-extinct 
Coenopteridales. The earliest of these ' Primofilices ' were only 
partly distinct from their presumable psilophytinean forebears, but 
others soon became characteristic and prominent elements of the 
flora. Further groups arose towards the end of the Palaeozoic era 
and persist to the present day — in the case of the relatively primitive 
ones such as the Marattiales apparently in decreasing numbers, 
but in the case of the more orthodox and modern Filicales still 
plentifully. Many even in early times were much like those, living 
nowadays, that are described in Chapter II and illustrated in Fig. 17. 

Fossil Seed-Plants 

Probably more important than true Ferns in the Carboniferous, 
and certainly accounting for a large proportion of the fern-like 
foliage of that and the following periods up to the mid-Jurassic, were 
the Pteridosperms or Seed-ferns, also called Cycadofilicales. As 
their names imply, these were fern-like (often tree-fern-like, but with 
secondary thickening) in some respects. But they bore seeds, as may 
be seen from Fig. 33, which shows a reconstructed plant and part 
of a frond of different types. The seeds were more or less naked, 
this group belonging to the Gymnosperms ; but although there are 
abundant differences, there remain sufficient deep-seated similarities 
to suggest that they may have given rise to the groups dealt with in 
the next paragraph, even as they themselves probably arose in pre- 
Carboniferous times from Ferns or fern-like stock that had not 
advanced far beyond the psilophytinean stage. Apparently allied 
or belonging to this group are the Caytoniales, which in some respects 


suggest primitive Angiosperms. The Gnetales may perhaps have 
sprung from the same stock ; but such fossils of them as are known 
are relatively modern, and so the origin and relationships of this 
group remain obscure (in Fig. 37 it is left near the Angiosperms). 
The Mesozoic has often been called the era of Cycads, owing to 
the presence in its deposits of some Cycads and of many more 

A B 

Fig. 33. — Pteridosperms (reconstructed). A, Lyginopteris oldhamia (after Berry) 
(X about To); B, Sphenopteris tenuis, portion of leaf with seeds (after Halle) 

(x f). 

Cycad-like Gymnosperms (Cycadeoidales) which between them com- 
prise the Cycadophytes. The living Cycads (an example is illus- 
trated in Fig. 18, A) are apparently the relics of a once more important 
and widespread group, whereas the Cycadeoidales have long been 
extinct. By many authors the Cycadeoidales are kept as a separate 
group, also called the Bennettitales, on account of their remarkable 
reproductive structures that were apt to look more like flowers than 
cones (a reconstructed example is shown in Fig. 34, A) ; others were 
slender and branching but probably never very big. The Cycado- 
phytes presumably arose from pteridospermous ancestors during the 
late Palaeozoic, the living Cycads supposedly representing the end 
of a line that in most respects has changed little since the early 



T-V'.". 1 ' 




Fig. 34. — A reconstructed Cycadeoid and parts of living and fossil Ginkgoales. 

A, Cycadeoidea, showing the flower-like strobili on the squat stem (x about %o)\ 

B, branch of Ginkgo with seeds ( X f) ; C, leaves of two species of Baiera, Mesozoic 

Ginkgoales ( X about f). 

Mesozoic, when its members were far more (probably very) wide- 

Of comparable antiquity, extensive development in the Mesozoic, 
and bare persistence to the present day are the Ginkgoales, repre- 
sented among living types by only the Maidenhair-tree, Ginkgo 
biloba. This is very restricted in anything like a wild state, but 
is widely familiar in cultivation ; a female sprig is shown in Fig. 34, B, 
as also are leaves of relatives which were almost world-wide in the 
Mesozoic. Like the Cycads, Ginkgo has motile sperms ; its obviously 
primitive characters have led to its being called a ' living fossil '. 

An earlier, long-extinct and apparently quite distinct group of 
Gymnosperms, the Cordaitales, flourished chiefly in the later part 
of the Palaeozoic, constituting with the Pteridosperms the bulk of 
the seed-plants of the Carboniferous coal-forests. Their origin 
is obscure, the earliest known members apparently being already far 
advanced. The Cordaitales were mostly large trees with sizeable 
flattened leaves and bearing their pollen and seeds in slender strobili 
as indicated in Fig. 35. 

The other major group of fossil Gymnosperms is the Conifers, 
whose surviving representatives afford the vast majority of living 
Gymnosperms. Examples of Conifers are illustrated in Fig. 18, 
their general features being described in Chapter II. They bear 
some resemblance to Cordaitales, and as fossils went back at least to 
the upper Carboniferous, an example from that time being shown 
in Fig. 36, A. Subsequently they appeared to reach their develop- 
mental climax in the Mesozoic, when several were similar in external 
form to those surviving today. Before the end of that era they began 


to tail off, some groups becoming less diverse and more restricted in 
geographical range, though as a whole they remained fairly numerous 
as well as various and ecologically important. It is widely supposed 
that the Coniferales derived originally from cordaitalean stock, 

Fig. 35. — Cordaitales (reconstructed). A, various Cordaitales (about 20 metres 

in height); B, end of a branch of Cordaites with strobili and a young branch 

(middle, right) (x about \). (Both after Grand'Eury.) 

though we cannot yet be certain of this. And in spite of the aggres- 
sive nature of some familiar types — such as the Pines and Spruces 
which still dominate huge tracts, especially of the colder lands — it 
seems that many others are eking out a rather precarious existence. 


Thus the group as a whole appears to be retreating before the angio- 
spermic onslaught , from which it has suffered since before the dawn 
of the Cainozoic era. Much as in the case of the so-called Pterido- 
phytes, it may be that the Gymnosperms represent several largely 
separate stocks as far back as we have yet any knowledge of them. 
The Angiosperms, as we saw in Chapter II, are the most highly 
evolved and successful group of plants at the present time, affording 
most of the dominant species on land. After an apparently slow 
start at least as early as the Jurassic, they gave rise before the end of 

Fig. 36. — Fossil parts of Conifer and Angiosperm. A, branch of Lebachia 

( Walchia) frondosa, one of the Palaeozoic Coniferales (after Renault) ( X about |) ; 

B, leaves of a dicotyledonous Angiosperm ( X about ^). 

the Cretaceous to a vast assemblage of forms that between them 
rapidly came to cover practically the whole surface of the earth, 
comprising most of the land-vegetation we know today, and a good 
deal of that developed in water. A considerable proportion of the 
genera and even species which are familiar to us nowadays as common 
shrubs or trees, particularly, seem to have persisted throughout the 
Cainozoic — often with little if any apparent change, and sometimes 
doubtless as dominants. Of the two main groups of Angiosperms, 
the Monocotyledons are much less numerous as fossils, and especially 


as early fossils, than the Dicotyledons. An' example of these last is 
shown in Fig. 36, B. 

Altogether the Angiosperms seem to be still in their early stages 
of expansion, with their fossils not yet expressing clear develop- 
mental trends. As to their future evolution we can only conjecture, 
and in regard to knowledge of their origin we are scarcely better off ; 
for each and every one of the gymnospermous groups, and in addition 
the Ferns, have been suggested by different authorities as having been 
their precursors. The outcome has been to stress our ignorance and 
leave the question unanswered, as we know of no series of fossil 
forms connecting the flowering plants with more primitive groups. 
Yet it does seem probable that they evolved from some primitive 
unspecialized group rather than from any modern and familiar one. 
At present it appears that if precursors of the Angiosperms are ever 
to be found, it will most likely be in the early Mesozoic or just possibly 
the late Palaeozoic, and that the Pteridosperms afford the most likely 
source of such stock as, conceivably, may in due course have evolved 
into the Angiosperms we know. This speculation arises from the 
facts that the Pteridosperms do not seem to constitute such a ' dead 
end ' as the other groups of Gymnosperms, whether living or extinct, 
and that some of their members or allies (the Caytoniales) have their 
seeds largely enclosed in a recurved cupule which is suggestive of the 
arrangement in the angiospermous fruit. As pointed out by Dr. 
H. Hamshaw Thomas (in litt.), the possible connection of the Angio- 
sperms with the Pteridosperms is further suggested by the similar- 
ities of structure in (a) wood anatomy (homologous types), (b) leaf 
form, for example in Glossopteris and Gigantopteris, (c) male flowers, 
and (d) seeds, especially as regards Angiosperms with integumentary 

Past Ages and Their Plant Life 

Fig. 37 aims to show the distribution of the main plant groups in 
geological time, with some suggestion of their relative abundance as 
far as this is known and can be indicated by varying the thickness of 
the figures representing the respective groups. It also indicates the 
geological eras and main periods, etc., mentions in sequence the 
dominant forms of life, and gives a series of supposed ages (cf. 
pp. 128-9). 

We will now proceed to a brief review of the main floras of the 
past, having familiarized ourselves with the sequence of geological 
time and with the chief groups of plants concerned. 

f 2 


S| o 



q-s: z 












„"- *< o 




The pre-Cambrian and early Palaeozoic preceding the Devonian 
together represent probably more than nine-tenths of known geologic 
time and constituted the age of Schizophytes, Thallophytes, and 
invertebrate animals. Bacteria, Blue-green Algae, Fungi, and true 
Algae of various groups appear to have been the main plants living 
through most of this time, although towards the end of it, in the 
Silurian, there seem to have been added the first land-plants in the 
form of some primitive Psilopsida and Lycopsida, and the still more 
problematical Nematophytales. 

The Devonian and Carboniferous periods constitute the real age 
of Pteridophytes and Fishes, even if these groups had an earlier 
origin. Thus in the Devonian the primitive groups already men- 
tioned apparently continued to flourish, as have Bacteria, Fungi, and 
Algae to the present day, while there were added various Sphenop- 
sida, etc. Apparently none of these early Pteridophytes has sur- 
vived, at least at the generic or lower level, any more than have the 
primitive Ferns which first appeared in that period, or possibly 
the Gymnosperms which were well established before the end of the 
Carboniferous. Although the early Devonian is often spoken of as 
primarily psilophytinean, there appear to have been plentiful other 
Pteridophytes living at the time. These evidently became more 
diverse and numerous in the upper Devonian, when the Psilophytales 
seem to have died out, and by then to have included some of tree 
dimensions. These may show what appear to be annual rings and 
indicate marked seasonal changes. An attempted reconstruction of 
a late Devonian forest is shown in Fig. 38. 

The Carboniferous period was the culmination of the age of 
Pteridophytes, including many of large tree form as indicated in the 
diagrammatic reconstruction (Fig. 39). To these the Lycopsida, 
Sphenopsida, and Filicineae all contributed, and, in addition, the 
early gymnospermous Pteridosperms and Cordaitales. The Bryo- 
phytes first appeared at this time, at least so far as known indubitable 
fossils are concerned, as did several of the modern groups — including 
the true Club-mosses, Horsetails, and some Ferns — while before its 
end there existed some primitive Conifers. The upper Carboni- 
ferous or Pennsylvanian was the great Palaeozoic coal age, and 
evidently a time of damp and widelv favourable ' even ' climate in 
both northern and southern hemispheres. The vegetation which 
gave rise to the immense coal deposits of the northern hemisphere 
seems to have been one of the most widely luxuriant of all time, 
though the coals of the southern hemisphere appear to have been 



formed from the remains of a flora which' was relatively poor in 
species and almost certainly belonged to a later age. 

Of late Carboniferous time the early Permian was essentially a 
continuation, though, later, arid conditions became widespread, the 

Fig. 39. — Generalized reconstruction of a Carboniferous forest. The tall, much- 
branched trees on the left are Lepidodendron. Below these are Sigillaria and leaves of 
Lyginopteris. In the centre foreground is a slender type of Calamite and in the back- 
ground are much-branched tree types. To the right of the water in the foreground 
are more Sigillarias and a Tree-fern. On the extreme right are lofty Cordaitales, 
and, in the undergrowth, several Pteriodosperms, among which is seen a SphenophyUum. 

big Pteridophytes and the Pteridosperms tending to decline and 
become replaced by other groups — including the drought-resisting 
Cycadophytes and more Conifers. At the same time the character- 
istic if limited Glossopteris flora existed under the relatively cold 
conditions of the south, in the areas of India, Africa, South America, 
Australasia and Antarctica that were supposedly occupied by the 
ancient ' Gondwanaland '. For the Permian was a period of active 
mountain building and rearrangement of large areas of land and sea, 
with severe glaciation at least in the South. It may be considered as 
starting the real age of higher Gymnosperms, which held sway 
throughout the Mesozoic era. In contrast with the generally uniform 
growth of trees in the coal age, those which grew during and immedi- 


ately following the Permian glaciation show strongly developed 
' annual ' rings. 

The early part of the Triassic period, at the beginning of the 
Mesozoic, tended to have an arid and generally unfavourable climate 
in the northern hemisphere, the fossil record being, moreover, 
fragmentary. Pteridosperms, Cycadophytes, Ginkgoales and Coni- 
fers appear to have been plentiful, as well as numerous Ferns and a 
lesser number of Lycopods and Horsetails. In Argentina, South 
Africa, and much of southeastern Asia, the climate was humid at 
that time, and supported a rich flora. It became drier and probably 
arid in these regions towards the close of the Triassic, when to the 
north conditions became much as in the Jurassic. Nevertheless the 
wide occurrence of similar floras indicated that comparable con- 
ditions extended over an extremely wide area. Fig. 40 shows an 
attempted reconstruction of a Triassic landscape, with plentiful 
Cycadophytes, etc., and some large Horsetails. 

Jurassic floras were apparently developed under warm and moist 
conditions as indicated by fossil deposits, and were widely distributed. 
Indeed, their composition appears to have been fairly uniform all 
over the world, involving (with relatively minor variations) such far- 
flung lands as continental Europe, Spitsbergen, Greenland, tem- 
perate North America, Mexico, India, Japan, Australia and New 
Zealand. They included numerous Cycadophytes, Ginkgoales, and 
Conifers, besides representatives of most of the modern groups of 
Pteridophytes, though the giant Lycopods and Horsetails had long 
since disappeared, as had the Cordaitales. During this period the 
Pteridosperms declined and, according to some authorities, the first 
indubitable Angiosperms appeared. 

The Cretaceous period witnessed the latest transformation of the 
plant world, in which the Angiosperms really came into their own. 
Conversely, most of the other groups of vascular plants were on the 
decline. It is interesting to note, however, that the lower cryp- 
togams had meanwhile often held their own, as many do to this day — 
presumably because they do not normally compete with Angiosperms, 
though actually they have not been widely dominant on land since 
early Palaeozoic times. Thus, of the widespread upper Cretaceous 
' Dakota flora ' of North America, more than 90 per cent, of the 
species were Angiosperms, most of them belonging to familiar woody 
genera of Dicotyledons, although the presence of a few Palms 
indicated a fairly warm climate. Cycads were reduced to 2 per cent, 
and the Conifers were only slightly better represented. Whereas 




apparently still more warmth-loving floras appeared during the 
Cretaceous even in the continental United States, the existence at 
that time in Greenland and other arctic regions of floras of temperate 
or warmer type suggests that the climate was widely genial, and again 
probably of comparable nature over most of the world. 

Fig. 40. — Reconstructed Triassic landscape (after Heer). 

In the early part of the Cainozoic, individual species of plants 
tended to be more widely distributed than they are today. Of the 
first five parts of the era, making up the Tertiary, the second, 1 or 
Eocene (by many considered the first that is really distinguishable), 
affords widely distributed floras indicating still favourable conditions. 

1 The Paleocene is nowadays frequently distinguished as the first, especially 


Thus in North America and across Eurasia, conditions were sub- 
stantially more favourable than at the present time, with Palms 
extending plentifully into Canada and England. Already the flora 
had a largely modern aspect, though herbaceous species in general 
and Monocotyledons in particular appear to have been far less 
abundant relatively to woody ones than at present. 

There may have been some deterioration of climate in the northern 
hemisphere in the next age, the Oligocene, but it appears to have 
been slight, with plentiful large trees prevailing well north. In 
any case in the southern hemisphere the climate seems to have 
remained more comparable with that of today than it did in the 
northern hemisphere. 

The Miocene was a time of widespread volcanic activity in North 
America, when uplifting of the Cascade Range deprived the area to 
the east of much of its accustomed rainfall, so that increasing aridity 
prevailed — and probably lower temperatures, although the climate 
was still warm over wide areas. Fig. 41 shows a reconstructed scene 
in Miocene times, with Palms and Cycads growing in what is now a 
cool-temperate region. 

By the Pliocene, conditions in America east of the Cascade Range 
had become generally unfavourable for the growth of dense forests 
and for the preservation of their remains, trees apparently occurring 
chiefly along the streams. As a result of the general cooling in the 
North, the vegetation became more like that of today, and actually in 
North America and eastern Asia the Pliocene deposits contain a 
large proportion of the species still found living in the same regions. 
On the other hand, many of the wide-ranging species then occurring 
in Europe have since disappeared therefrom. 

In spite of all these changes in various parts of the Cainozoic, the 
same major plant groups appear to have persisted through it to the 
present day, even if the species and often higher taxa have changed ; 
particularly striking is the dwindling importance of some of the 

The Pleistocene and Recent together make up the Quaternary and 
are now estimated to involve only the last million or fewer years 
(cf. p. 129), of which the Recent or post-Pleistocene occupies per- 
haps one-hundredth part. Remains of Pleistocene vegetation are 
preserved chiefly in unconsolidated lake and stream deposits, in 
peat bogs, or in a frozen condition, while postglacial peat deposits are 
still accumulating. In general only modern species are represented, 
most of these being still living and familiar ; but in Pleistocene 



deposits they often extend far (and in post-Pleistocene ones sometimes 
considerably) beyond their present limits, suggesting substantial 
changes in climate since they were laid down. Indeed, such changes 
we know to have taken place in the northern parts of Eurasia and 

Fig. 41. — A reconstructed scene in Switzerland during 
Miocene times (after Heer). 

America, where several periods of advancing and declining glaciation 
widely affected the climate, and where further changes are indicated 
by such evidence as that afforded by sub-fossil pollens and other 
remains of contemporary plants preserved in bogs. These remains, 


when identified with plants of known climatic requirements, afford 
a valuable indication of local conditions. And as a further instance 
of the significance of climatic change, we shall see in the next 
chapter how the great diversity of woody plants in North America 
and eastern Asia is attributable to the fact that in these regions 
such plants were able to migrate far south before the extending ice 
and return north after its margin had receded, whereas in Europe 
they were supposedly forced against the southern mountains or sea 
and exterminated. For such reasons we must examine in the next 
chapter the historical bases of modern geographical ranges before 
we can deal in an understanding way with the distributions we 
actually find. 

Further Consideration 

The facts and theories advanced in this chapter, as in the case of most 
others, have usually been gleaned from various specialist and often highly 
technical sources which the general reader will scarcely wish to consult 
even if they are available to him. However, further details (and some- 
times other opinions) may readily be obtained from one or more of the 
following generalized or introductory books in English : 

Sir A. C. Seward. Plant Life Through the Ages (Cambridge University 

Press, Cambridge, Eng., pp. xxi -\- 601, 1931). 
C. A. Arnold. An Introduction to Paleobotany (McGraw-Hill, New 

York & London, pp. xi -f- 433, I 947)< 
H. N. Andrews. Ancient Plants and the World They Lived in (Comstock, 

Ithaca, N.Y., pp. ix + 279, 1947). 
John Walton. An Introduction to the Study of Fossil Plants, second 

edition (Black, London, pp. x -f- 201, 1953). 

Chapter VI 


The distribution of each kind of plant making up a modern (and 
presumably any other) flora is apt to depend upon (i) the history 
of the plant in geological and recent times, (2) what may be called 
its migrational ability, and (3) its adaptability in physiological and 
other ways to the conditions of such new environments as it may 
reach. To a considerable extent migrational ability depends upon 
the efficiency of dispersal (as described in Chapter IV), and adapta- 
bility on plasticity of form or function (as treated in Chapter III). 
This leaves the ' fossil history ' and recent vicissitudes to be dealt 
with next, at least in such aspects as are best known or most pertinent. 

Whereas we have seen that, apart from such periods of local or 
regional change as occurred most notably in the Permian, the flora 
and vegetation of the aerial parts of the globe tended to be widely 
comparable in different regions during the earlier geological ages up 
to and including the Mesozoic, this relative uniformity was not 
maintained through the Cainozoic. There is plentiful evidence to 
show that still later than the end of the Mesozoic, during early 
Tertiary times, the forests tended to be more widespread and more 
uniform than at present — practically throughout the land of the 
northern hemisphere, including what are now high-arctic regions. 
And although climatic belts doubtless existed during these and earlier 
times, they can scarcely have been as marked in favourable periods 
as those we know nowadays ; for, with luxuriant vegetation flourish- 
ing within a few degrees of both the North and South Poles, the 
tropics would have been too hot for normal life — at least if the land 
of the world had at all the conformity it has today. But particularly 
from the Miocene onwards there were marked local changes in 
conditions, and so these latest fossil and sub-fossil floras are of great 
significance to us when considering the problem of the origin of the 
flora existing today. 


foundations of modern distributions 155 

Some Effects of Relatively Recent Climatic Changes 

At the close of the Tertiary or beginning of the Quaternary, 
although the vegetation of the tropical and adjacent zones continued 
in considerable luxuriance, there was a marked lowering of tempera- 
ture in most other regions, that led to the covering of some by 
glaciers and to complete changes in the floras of others. Climate 
being the primary controller of vegetation, the flora in favourable 
warm areas probably continued, as it does to the present day, much 
as it had done at least through the middle Tertiary. But in the 
less favoured belts to the north and south, a lowering of temperature 
and decrease in precipitation — for instance in much of the Mediter- 
ranean Basin, in eastern Europe, and in northern and central Asia 
— led to the widespread replacement of the warmth-loving floras 
by hardier types. It may be presumed that there was at the same 
time a reduction in numbers of species and individuals, and in the 
general luxuriance of the vegetation. And whereas in the boreal 
regions, and even in the Arctic, there had formerly flourished (and 
according to some authorities evolved) a vast array of warmth-loving 
or at least mesothermic {i.e. liking moderate temperatures) types, 
these were in time pushed far south or, in some cases, doubtless 

Although evolution in general is a gradual process, it seems to 
be accelerated by sudden changes in habitat conditions. Thus the 
violent upheavals of the earth's crust or direct changes of climate 
appear to have induced sudden inherent and hereditary changes 
involving the abrupt creation of new races and, in time, of new 
species. Contemporaneously and for similar reasons, floras have 
been caused to migrate. These variations of conditions over long 
periods of time have probably led to a far more intricately adjusted 
and highly evolved general flora and vegetation than would have 
obtained if conditions had remained as they were, for example, in 
the Carboniferous. Indeed, some authorities have pictured a world 
of gigantic Reptiles instead of Mammals, and of giant Club-mosses 
and Horsetails instead of Angiosperms, as continuing today if the 
Carboniferous climatic and other conditions had persisted. 

Besides direct climatic changes and others depending on such 
geological revolutions as the thrusting up of mountain ranges, there 
have evidently been changes in the conformation of land and sea 
— even, according to some authorities, in the positions of the con- 
tinents with regard to one another and in relation to the geographical 


poles. All such changes would inevitably lead to alterations in the 
distributions of plants and of the animals that are dependent upon 
them, profoundly affecting also their evolutionary development. 
Thus, whereas in the [northern hemisphere the distribution of the 
climatic zones favoured the development of relatively luxuriant 
vegetation over wide areas up to nearly the end of the Tertiary,] in 
the southern hemisphere conditions appear to have been much more 
disturbed. Particularly did the great Perjmian gkciation, whose 
principal centre of development apparently lay in South Africa, play 
havoc with the flora and fauna ; to it is attributed in some degree 
the poverty to this day of the flora of tropical Africa as compared 
with the floras ofSouth America and Asia. It is supposed that 
relatively arid conditions prevailed subsequently over much of the 
southern hemisphere, which may have been further diversified by the 
separation of the continents beginning as far back as the Mesozoic. 
Thus tropical Africa is supposed to have been cut off from the rich 
vegetation of Eurasia by a wide sea occupying the position of the 
Sahara during the JJretaceous, and, in later times, by the Sahara 
Desert as we know it today — both sea and sand being effective in 
barring the migration, it has been suggested, of the majority of 
species from the North. 

The climatic deterioration that began well back in the Tertiary 
led ultimately to widespread glaciation in the northern hemisphere. 
Thus, early in the Quaternary, vast areas of North America and 
northern Eurasia became enveloped in ice which may be compared, 
in extent and continuity if not in thickness, with that covering most 
of Greenland and Antarctica today. This Pleistocene ice reached a 
maximum extent and receded probably four times, the limits reached 
in each extension being different. The intervening, ' interglacial ' 
periods were protracted and relatively favourable to plant life, being 
apparently for long periods at least as warm as the present day. 
The ice largely ousted the previously rich floras of the North, the 
plants being forced to migrate before its advance or, if they could 
not do so successfully, being exterminated — apart perhaps from some 
which persisted on mountains or other ice-free refugia, a subject 
to be discussed in the next section. 

The poverty of the present western and central European flora 
is commonly ascribed to the fact that the Tertiary components were 
largely destroyed shortly before or during the Ice Age, by being 
driven into the sea or against high mountains. For post-Pleistocene 
restocking across the Mediterranean, over the Alps or Pyrenees, or 


from the floristically poor regions lying immediately to the east, was 
problematical to say the least, and from the west was wellnigh 
impossible as the Atlantic was now indubitably in existence. On 
the other hand, in North America and eastern Asia the warmth- 
loving plants among others were free — and evidently often able — 
to migrate far south in the lowlands or along mountain ranges, 
subsequently to return when the ice retreated and suitable conditions 
were reinstated in the North. Much the same appears to have 
happened in the Balkan region. To such considerations is attributed 
the presence of many living woody plants, such as members of the 
Magnolia family (Magnoliaceae), in eastern North America and 

Fig. 42. — Known distribution of species of Liriodendron in Tertiary times (black) 
and nowadays (hatched). (After Good). 

southeastern Asia but, on the other hand, their absence from Europe, 
though their remains in fossil deposits indicate that they were once, 
widely distributed and locally plentiful in all three of these continents 
Fig. 42 shows the known distribution of Tulip-trees {Liriodendron 
spp., Magnoliaceae) in Tertiary times (black) and living today 
(hatched), and Fig. 43 shows the even more drastic restriction of 
the giant Redwoods. Fig. 44 indicates the periods during which 
the various parts of North America are supposed to have been free 
from major ice-sheets. Even in the last of the interglacial periods, 
many plants persisted much farther north than they grow today — 
most notably in Europe — and it has been suggested that we have 
still not emerged permanently from the Ice Age. 

Whereas it is thought that considerable expanses of territory in 



northern Asia and smaller ones in northern continental North 
America, as well as the northernmost insular tracts, were free from 
major ice-sheets throughout the Pleistocene, owing for example to 
their ' continental ' climate and especially low precipitation, it is 
believed that ice at one or more stages of the Pleistocene engulfed 

FlG. 43. — Past and present distributions of Redwoods. Above, known localities 
of fossil Redwoods (after Chaney); below, modern (relic) areas of Coastal Redwood 
{Sequoia sempervirens) on left and of Sierra Redwood or Big-tree (Sequoiadendron 
giganteum) on right. (Modified from Cain : Foundations of Plant Geography, copy- 
right 1944, Harper & Brothers.) 

much of northern North America (see Fig. 44) and virtually the 
whole of western and central Europe as far south as the Thames 
Valley in the west and the Carpathians in the east. Nor could such 
extensive glaciation of northern territories obtain without being 
reflected in a lowering of temperatures to the south and even in 
the tropics, where some mountains became glaciated, and where 

j y k B,%\ Existing glacial fields. 

2 Exposed after disappearance of PLEISTOCENE ice. 
| <JZ. -j Driftless and nunatak areas wholly or partly exposed during PLEISTOCENE. 
r&g^A Areas of mountain and valley glaciation during the PLEISTOCENE. 

Atlantic coastal plain and Pacific areas exposed during the QUATERNARY. 

K] Probable land area available since close of the T E R T I A R Y . 

Probable land area available since close of the CRETACEOUS 
Probable land area available since close of the JURASSIC. 

_J Probable land area available since close of the P A L A E O Z I C , or unknown. 

Fig. 44. — Map showing periods since when various areas of present-day North 

American land have supposedly been free from major ice-sheets, etc. (Harvard 

University handout, modified after Flint and others.) 



warm lowlands (at least in the southern hemisphere) experienced 
' pluvial ' periods of increased precipitation. Simultaneously there 
was more extensive glaciation in some regions of the South, notably 
the uplands of South America and southern Australasia. All these 
and allied changes w r ere gradual, the advance and retreat of the ice 
each time occupying many thousands of years, and allowing plants 
to migrate before its face in accordance with climatic changes. 

Numerous living genera, especially of woody plants, are known 
once to have been far more widely distributed than they are today 
(striking examples are shown in Figs. 42 and 43). Many of our 
species and even vegetational associations are of similar nature to, 
but geographically more restricted than, those developed as far back 
as the Miocene. It seems that by this time most of the major 
land-masses of today had been formed, at least in outline, and that 
thereafter the deterioration in climate was marked, leading to a 
considerable reduction of the areas of many species of plants and 
animals — quite apart from the restriction effected by or through the 
Pleistocene glaciation. Already in the Pliocene the floras had 
tended to be markedly impoverished as compared with those of the 
Miocene, considerably less than 1,000 species of plants being known 
from the Pliocene in contrast with over 6,000 from the Miocene. 
The analogy of animals would suggest that over 80 per cent, of 
Pliocene plant species are still living today — some of them, e.g. the 
Bog-cypress (Taxodium distichum) and Black Oak (Quercus nigra) 
in Alabama, in the selfsame region. In other instances they are 
not now known to live in a wild state within thousands of miles of 
their old haunts — as in the case of the Water-chestnut (Trapa), fruits 
of which are widespread in deposits in North America up to the 
Pliocene, though it appears to be no longer a native of the New 
World. An interesting set of examples is afforded by a middle 
Pliocene flora of western Europe whose closest agreement is to be 
found with some areas of southeastern Asia ; on the other hand, 
the flora yielded by the overlying late Pliocene beds finds its closest 
relationship with the existing flora of central Europe and indicates 
an already much cooler climate. 

It is supposed that the Miocene and early Pliocene floras were 
practically circumpolar in distribution and also widespread latitudin- 
ally. Then the increasingly colder conditions in the North, often 
accentuated locally by the uprising of mountain ranges which 
furthermore caused aridity in their rain-shadows, forced the plants 
to migrate_southward. It has been suggested that in this pre- 


Pleistocene southward migration there were three principal avenues 
of escape, determined primarily bx the_pr^yajlin^--4ir^c4k)n- of the 
mountain systems : (i) along the lowlan ds of eastern Asia, (2) along 
the mountain ranges of North America^ and (3) down the Scan- 
dinavian Peninsula and adjacent areas into western and central 
EuTope. The first route may have led to i^rmixturfi_£>£narthern 
species with the original native flora ^ and _ so may -largely --explain 
the richn^^ofjhe existing flora of China. The s econd route may 
have had much the same effect i n Ame rica and would, moreover, 
explain the similarity, between the floras of eastern North America 
and eastern Asia. In the third instance, however, the great European 
mountain ranges with ^x^Jyjrig^e^sXBJidja^e^tJffiiiuM have, prevented 
fur ther s ^udrward^^migr-ation, with resulting impoverishment both 
currently and as regards possible recolonizing elements even after 
the final Pleistocene recession, as we have already seen. However, 
there has been some opposition to this hypothesis, and serious 
doubts expressed, for example, as to whether southwestern France 
was cold enough at this time to cause the postulated extermination, 
though this might still have been effected by the persistence there 
of closed communities, which are among the toughest barriers for 
a migrant to cross. Nor is it known whether the great extinction 
of plants and animals in Europe took place before the end of the 
Pliocene or during the Pleistocene. Probably it was a gradual pro- 
cess, each advance of rigorous conditions leading to the loss by the 
flora of some of its less hardy elements, so that potential waves of 
recolonizing vegetation became successively more impoverished. 

Pleistocene Persistence versus Subsequent Immigration 

The Pleistocene itself, as we have seen, was a climatically unsettled 
period of cold spells with extensive glaciations, which were separated 
by relatively long warm intervals (interglacials). In general the 
plants living during the Pleistocene were specifically identical with 
those of the present time, the chief differences exhibited by them 
being in spatial distribution. During some at least of the inter- 
glacials the climate appears to have been similar to that of the 
present day — which indeed has led to the suggestion that we may 
nowadays be merely in another interglacial. Thus more than 
70 per cent, of the species whose remains have been identified from 
an interglacial deposit in Germany live in the same district at present, 
and so do most of those in a deposit near Toronto in southern 


Canada, though a few are now restricted to areas slightly farther 

The climate near the margin of the ice during the glacial stages 
was probably rather like that of southern Greenland today, where 
arborescent growth is to be found within a few miles of the terminal 
face of the ice-cap ; even more striking is the situation in New 
Zealand, where subtropical Tree-ferns may be seen growing a bare 
mile from the end of a glacier. Thus the Pleistocene tundra (treeless 
plain) in North America probably formed only a narrow belt around 
the margin of the ice-sheet, with conditions becoming rapidly more 
equable away from it. In Europe the ice-free zone appears to have 
been relatively wide, at least during the later glacial maxima. In 
central Eurasia the tundra was apparently bordered by steppes, while 
in North America extensive deposits of loess (wind-transported 
sediment) suggest a drier climate than now prevails. To the south 
of the tundra and steppe belts lay more or less broken forest, although 
it seems likely that in some places the advancing ice impinged 
directly on the forests. When precipitation decreased or the climate 
became warmer, the ice retreated and the whole series of vegetation 
zones followed its margin northwards — only to be pushed south 
again when conditions deteriorated and the ice advanced once 

It is often supposed that in Alaska and eastern Asia the glaciation 
was chiefly of the local, mountain type, considerable areas being left 
free or at least not covered by extensive ice-sheets. It also seems 
that, because the precipitation was locally insufficient for extensive 
accumulation of ice, the northernmost parts of Greenland and 
the Canadian Arctic Archipelago remained largely unglaciated. 
There is no reason to doubt that many plants persisted in these 
unglaciated areas — in some cases probably throughout the whole of 
the Pleistocene, in others at least through its latest glacial maximum 
(which was by no means everywhere the greatest in extent as com- 
pared with its predecessors). 

But whereas it is an observation of today, and a supposition for 
similar phenomena of the past, that many plants can and do grow 
on ice-free areas even when these are surrounded by ice, it is another 
matter to explain apparent anomalies of modern distribution in terms 
of such survival throughout long periods of intense glaciation. Yet 
this has been so widely assumed that it seems necessary to point 
out the seemingly overwhelming counter-arguments to this ' nunatak ' 
hypothesis — so called from the Eskimo word for a mountain sticking 


out of an ice-cap, for on such mountains much perglacial persistence 
is supposed by some students to have taken place. 

To begin with, much of the supposed evidence brought forward 
for survival, such as the ' relict ' nature of particular plants in 
particular areas, has proved to be either capable of other interpretation 
or actually erroneous. Thus, geologists have insisted that several 
of the areas claimed by the advocates of persistence to have been 
1 refugia ' during the Pleistocene, were in fact glaciated. Moreover, 
latterly there have been numerous instances of plants of supposed 
isolation or disrupted distribution (which were claimed to indicate 
such refugia) being found in intermediate positions — sometimes even 
on islands that have only recently risen out of the sea. Nor does 
endemism (restriction to particular geographical areas) necessarily 
indicate isolation and persistence ; it sometimes actually suggests 
the opposite, as in some cases of hybrid origin ! 

It is noticeable that many of the isolated or restricted (endemic) 
species, which are made so much of by advocates of the nunatak 
hypothesis, are characteristic of, or sometimes restricted to, cal- 
careous soils. It has been strongly counter-claimed that their spotty 
or localized occurrence can best be correlated with soil characteristics, 
the presence or absence of signs of recent glaciation being of little 
or no consequence. And when we remember that many of these 
are far-northern plants which favour ' open ' soils where competition 
is lacking, and that in boreal regions calcareous areas are notable 
for their poor vegetation but diversified flora, other possibilities 
spring to mind and the nunatak hypothesis becomes less and less 

There are also objections to the nunatak hypothesis on the ground 
of far easier and wider dispersal than its adherents will admit — for 
example by Birds, and as regards characters of higher plants trans- 
ported in pollen, and through airborne disseminules of lower plants. 
Thus in the manner of plants, Birds also have their habitat pre- 
ferences, often traversing great distances from one to another similar 
spot. Although normally they are supposed to * fly clean ' when 
on migration, they must surely (as already mentioned in Chapter IV) 
sometimes carry materials frozen or otherwise stuck to their plumage, 
etc. — especially if flushed unexpectedly or unwell, when they are 
apt to ' neglect their toilet '. Pungent instances seem to be afforded 
by the seeds, coated in protective regurgitate, that have been found 
adhering to migrant Birds on the subantarctic Macquarie Island, 
but apparently do not belong to any of the local species, as noted 


already on p. 1 14. And in the final analysis it must be remembered 
that a single disseminule in many millennia may be sufficient to 
' plant ' a species. 

But whereas the idea of certain plants persisting in tiny areas for 
hundreds of thousands of years, unchanged and unaffected by the 
coming and going of ice-ages and hordes of other migrants, is scarcely 
in accordance with our saner biological reflection, some persistence 
on ice-free tracts, for example through the last glacial maximum, 
may well have taken place. And restocking may indeed have been 
helped by plants surviving on temporary sheltered ' islands ' sur- 
rounded by ice, or on ice-free coastal strips — thence to recolonize 
surrounding areas on release from the bondage of glaciation. On 
the other hand, the plants surviving on mountain nunataks would 
be mainly arctic or high-alpine ones accustomed to rigorous con- 
ditions and unlikely to recolonize lastingly the surrounding plains 
on recession of the ice and marked amelioration of conditions. Thus, 
as was pointed out by Professor G. E. Du Rietz (in litt.), ' Scan- 
dinavian botanists have believed more in glacial survival in coastal 
areas than on nunataks ', and such an ' hypothesis of glacial survival ' 
would seem more reasonable ; nor are the criticisms of the nunatak 
hypothesis which apply to North America necessarily valid in 
northern Europe where conditions tend to be different. Many 
persisting plants, however, appear to become depleted in biotypes 
and, owing to a consequent narrowness of ecological amplitude, 
rather passive ; some become reactivated by cross-breeding fol- 
lowing release from glaciation, aggressively recolonizing surrounding 
areas from their vegetated ' islands '. Thus, although the nunatak 
hypothesis in its strictest form seems unsound and indeed unneces- 
sary, some residue of it may well be valid and still found useful, 
in spite of the fact that the recolonization of deglaciated areas 
appears in general to have taken place by immigration of plants 
that tided over the inimical period in distant (usually southern) 
areas. Thence the plants subsequently migrated, for the most part 
by gradual stages. 

Persistence through the Pleistocene, or at least through its latest 
phases, may also help to explain the existence of ' arctic ' plants on 
mountains far to the south, although in some cases migrating birds 
are probably responsible. In other instances, the similarity to arctic 
conditions evidently enabled properly acclimatized plants to retreat 
to the mountain tops, even as they were able to persist near the 
margin of the ice and follow it north. For in both these situations 


the plants were able to get away from the competition of ranker 
(but often less hardy) types. 

Altogether it seems most reasonable to consider the post-Pleisto- 
cene flora of those territories of the northern hemisphere which 
were freed from the last major glaciation as being made up of : 

(1) some elements which had with little doubt persisted in at least 
recent unglaciated ' refugia ', particularly in sheltered coastal areas, 

(2) elements spreading from sheltered refuges afforded particularly 
by mountain systems where the Ice Age did not have such catastrophic 
consequences as to destroy the characteristic Tertiary flora even if 
it did cause impoverishment, (3) probably numerous elements which 
migrated northwards from the territories bordering the southern 
extremity of the ice after it retreated, and (4) recent immigrants 
from afar (or at least from regions not drastically affected by the 
Pleistocene), whose establishment has been favoured, in areas recently 
freed from glaciation, by the local lack of competition from already 
closed (i.e. continuous) vegetation. These recent migrants were 
aided by natural means — wind, water, or animals — or by Man, 
through intentional or accidental importation, and many such migra- 
tions are still going on plentifully all the time. Examples are afforded 
by the considerable numbers of weeds that have recently become 
established in ' open ' areas, including inhabited parts of West 
Greenland, and the westward advance of Asian species into Europe. 

Continental Drift, Shifting Poles, Land-Bridges, etc. 

The present distribution of any given plant species is in part a 
reflection of the geological revolutions and climatic changes that 
have occurred in the world during the period of its existence as a 
species. In former sections of this chapter we considered the 
effects of climatic change in the past and particularly the significance 
of the Pleistocene Ice Age. It is now time to examine some other 
leading theories that have been advanced in an attempt to explain 
the current distribution of particular plants. 

Perhaps the most promising and plausible (though still highly 
controversial) theory in this connection is that of ' continen tal drif t ', 
which is often identified with the name of its principal proponent 
of recent times, the late Dr. Alfred Wegener. This * displacement 
hypothesis ' is based on the assumption that the present-day con- 
tinents once formed part of a single land-mass (Pangaea), or, accord- 
ing to a recent modification, two land-masses, whose continents 


started breaking and drifting apartLcinring the Mesozoic era and 
went on doing so until they came to reach the positions they now 
occupy. According to some advocates of the theory, this drifting 
is still proceeding and actually demonstrable in the case of Green- 
land. Such recent drifting might conceivably have left Europe and 
North America in contact with one another until the early Quaternary, 
and likewise Antarctica and South America may be presumed to 
have remained long in contact. The theory also accounts for 
considerable rearrangement of the climatic zones on land, as the 
drifting of the continents was supposed to involve changes in their 
several positions relative to the North and South Poles. This 
in turn would allow marked changes in the distributions of living 
organisms, and bring plants and* even entire floras to regions whose 
present climates are barely adequate for their growth. Still more 
spectacularly, it would allow fossil remains to drift with their 
enclosing land-masses to distant parts of the globe — and hence into 
entirely different climatic belts — and this could explain, for ex- 
ample, the presence of fossils of warmth-loving plants in the Far 

Although it is rejected by some geologists and, especially, geo- 
physicists, the theory of continental drift has won the ardent support 
of others — even if only for earlier geological times than those which 
allow most help to the biogeographer in theorizing about the anomalies 
he finds. To many of the botanists among these biogeographers, it 
seems to constitute the most plausible working hypothesis upon 
which they can base their suppositions as to the history of plant 
ranges. In one connection or another, the controversy rages back 
and forth and probably will continue to do so for a long time to 
come, though it may be noted meanwhile that the maps produced 
by proponents of the theory are extraordinarily suggestive, and 
would, one imagines, if analyzed statistically, indicate a very high 
degree of probability (cf. Fig. 45, A). However, geologists and geo- 
physicists are prone to put back the time of possible rift and major 
displacement of continents well before the beginning of the Mesozoic, 
when it would be of little if any help to plant geographers in their 
search for explanations of striking similarities of flora in certain areas 
that are now widely separated by oceans. For it is obvious that, 
as differences between species commonly involve the establishment 
of various genetically independent mutations (sudden changes), the 
chances that two isolated populations will evolve in exactly the same 
way are incalculably low, while convergence in every respect of 


previously dissimilar types is even more improbable. Accordingly, 
similarities across wide stretches of water are more popularly 
explained by suppositions of previous proximity or even contiguity, 
though, as we have already suggested, dispersal may be more effective 
than is commonly supposed, and, conceivably, responsible for some 
at least of the apparent anomalies of distribution. 

If it were accepted for Mesozoic and later times, the hypothesis 
of continental separation and drift would make unnecessary, or at 
all events less necessary, several of the following and other theories 
that are of interest in plant geographical considerations — though it 
may be confidently assumed that it will not be so accepted without 
far more conclusive evidence than has yet been presented. Mean- 
while it should be noted that this theory is supported by, or is in 
accordance with, a great many data on the migrations and inter- 
relations of different floras and plant groups in past geological ages, 
besides explaining many anomalies of plant distribution of the 
present day. On the other hand it offers no explanation of the 
similarities of the floras of eastern Asia and North America, or of 
the floras of islands in the Pacific which suggest trans-Pacific con- 
nections — or at least migrations. Indeed, according to Professor 
G. E. Du Rietz (voce), who yet believes it possible that the same 
taxonomic entity may have arisen independently in more than one 
area, the flora of New Zealand is contrary to the hypothesis of 
continental drift, bearing similar relationships to both its east and 
its west. 

The theory of polar oscillations or ' shifting of poles ', sojnetimes. 
called jthe^ Vp_enduJjLrrriJ±ieory_', also aims at explaining, on the -basis 
of incidentaL_£arlier climatic changes, some phenomena^ ojL plant 
distributiomthat are far out of line with climatic zones as they now 
exist. It is presumed that cjjgnge^Jn^c limatic ^zongs , as indicated 
inter alia by the distributions of fossil p lants, were cause d by changed 
locati on of the co ntine nts in relatk)n_to_jhe_suii!_s_ j)_rientatiojCL^_and 

by_agju ming that __perj odic changes__have occurred, owing to the 
position of the geographical poles oscillatin g-back-and ioith-like-a 
pendu2um J _or_at_ least— ^wandering ! quite w idely ^ This must not 
be confused with the now well-known movement of the North 
Magnetic Pole. It should be noted that continental drift and 
(geographical) polar wandering are considered to have quite possibly 
taken place together, in view of the plasticity implied by the former, 
and that proponents of the latter are prone to put an earlier position 


Upper Carboniferous 

(■■■■■ ( mwi^ 

\Jn ' 

i^OV ^sN® \ \ 



Cv\ \ \v 


Old Quaternary 




GREENLAND (827 300 so. miles) 

Fsowerin*. Plants 420 species (less chin 1% tndem 
Pteridophytw 20 species (none endemic) 

BRITISH ISLES (130.800 so. miles) 
ml, ,et be lound or> ine 


CETLON (25.300 so. miles) 
Ftowenne Fl.nts abou 
Ptenoopnyra ibov 

2.300 species 34% 
l 2S0 species lendemie. % 

JAMAICA (4.400 so. miles) 
Plendopnjtes ibosi 

2.500 sp 
523 sp 

ecries (.bout 20% 
ccies (.bout 12% 




Fig. 45. — Maps illustrating ' Continental drift ' and the proximity of land-masses 
as apparently affecting floristic richness. A, reconstructions of the map of the 
world at three stages according to the theory of continental drift, the thinly stippled 
areas being shallow seas, while present-day rivers, etc., are shown merely for pur- 
poses of orientation (after Wegener); B, comparison of areas and floras of islands 
in relation to the proximity of major land-masses and their floristic richness 
(prepared by W. T. Stearn). 


of the North Pole in the North Pacific, the implied change in the 
earth's axis bringing the South Pole into the South Pacific. 

If the theory of polar oscillations were accepted, it would in turn 
make unnecessary any separate theory of land-bridges, as changes 
in level of the sea induced by oscillations of the earth (that are pre- 
supposed to have caused those of the poles) would suffice to cause 
the joining together or separation of different parts of the earth's 
land surface. However, in spite of some recent attempts at resur- 
rection, and continuing discussion of the possible instability of the 
earth's axis and of the phytogeographical implications this might 
have, 1 this theory of polar oscillations is said to possess little geo- 
physical foundation or geological support. Moreover, if it were 
made to explain some anomalies of plant distribution, it would 
apparently merely precipitate others ! 

The theory of land-bridges has long been popular in some quarters, 
and indeed there are few seas or even deep oceans that have not 
been hypothetically bridged by one or another over-enthusiastic 
author to explain present-day anomalies in plant or animal distribu- 
tion. And certainly the simplest way to explain similarities (some- 
times only supposed) of plant and animal life between such areas 
as Europe and eastern North America, or Australia and South 
America, is to assume that they were once connected by a bridge 
of land or a ' lost continent ', though the bridging and hence pos- 
sibility of migration across the present-day ocean need not necessarily 
have been continuous at any one time. But altogether the theory 
strikes the sceptic as being too artificial and ' convenient ' for reality, 
and again there are contrary geophysical and geological arguments. 
It was advanced chiefly as a concession to some aspects of bio- 
geography, while leaving others unexplained. Thus it leaves 
unclarified why plants in former geological ages flourished in regions 
whose climates are now far removed from those to which their 
modern counterparts are accustomed, and it also gives no satisfactory 
explanation of discontinuous areas of distribution. For if the 
fioristic similarities of two distant areas having alike plants were to 
be explained by their having once been joined by a ' bridge ', it 
would be necessary to assume that the entire extent of that bridge 
offered similar ecological conditions suitable for the migration of 
these plants. This requires more imagination than to visualize 
recent dispersal by the means discussed in Chapter IV, or, perhaps, 

1 See, for example, Nature, vol. 175, p. 526, 1955, vol. 176, p. 349, 1955, and 
vol. 176, p. 422, 1955. 


continental drift involving the ' like ' areas being once together but 
subsequently separated ! 

It seems, however, that the theory of land-bridges is justified to 
the extent that such phenomena come and go nowadays on a small 
scale — for example with the emergence and submergence of isthmuses 
with changes in the relative level of water, and with the throwing 
up or destruction of beaches by the sea. Also, there can be little 
doubt that such ' bridges ' have existed on a bigger scale in the past, 
at least to the extent of once linking together some present-day 
islands with their adjacent mainland over an area of continental 
shelf, and joining in continuity such close land-masses as Alaska 
and easternmost Siberia. Nor, according to some authorities, can 
certain distribution-patterns in the southern hemisphere well be 
explained without the supposition of an Antarctic land-bridge. 
This may have joined South America to Australasia, to which two 
regions such restricted genera as Nothofagus (Southern Beeches) and 
Fitzroya are common, while also occurring as fossils on the Antarctic 

The theory of permanence of oceans and continents, which grew 
out of objections to, and largely opposes, those of land-bridges and 
continental drift, presupposes the land-masses to have occupied their 
present positions from pre-Cambrian times. However, it still leaves 
unexplained many biogeographical phenomena — particularly those 
that suggest the continents to have been connected at some period 
— and is unable to cope with the high-arctic flourishing of luxuriant 
vegetation in earlier ages. For although it has been suggested, and 
may yet be maintained, that plants might have changed their climatic 
requirements in the past, they can scarcely have done this sufficiently 
to account for such extreme cases. 

This brings us to the theory of the ,pillar_ongin_of ^floras, which 
in one or another form has some eminent advocates to this day. 
When it was believed that climatic conditions were uniform through- 
out the world in earlx geological ages, jwith ^differentiation into 
climatic zones not taking place until the end of the Cretaceous or 
beginning of the Tertiary, it was widely thought that the floras of 
the world spread rapidly from a single centre lying in the north 
polar region — through Europe into Africa, through eastern Asia into 
Malaysia_and ~Aiistr3ll5^ncLthro^ l^outh 

America. This was the arctic or monoboreal theory of the origin 
ol floras. Subsequently such discoveries as that of fossil floras in 
Antarctica indicated that a similarly drastic reduction in temperature 


has taken place there, and suggested that the lands of the Far South 
had once been connected with one another by way of the Antarctic 
Continent. This provided a basis for the presumption that there 
had probably been a centre of species-formation there too, and that 
life originated in the lands encircling both geographical poles. 
However, neither theory can well be accepted, for it is now recognized 
that climatic zones have existed as long as there has been life on 
earth, and that ice-ages occurred previously to the Pleistocene and 
in parts of the world other than the present polar areas. Moreover, 
other centres of plant development have been strongly advocated 
— particularly in the tropics. 

An interesting example of the importance of historical and other 
factors in the diversity and constitution of present-day island floras 
is illustrated in Fig. 45, B, kindly contributed by Mr. W. T. Stearn, 
of the British Museum (Natural History). This shows that whereas 
the flora of Jamaica is large and diverse although its area is small, 
the flora of Ceylon is smaller although its area is larger than that 
of Jamaica, while the flora of the many times larger British Isles 
is smaller still. Each island having a comparable degree of topo- 
graphic variation, and the climates of Jamaica and Ceylon being 
similarly favourable and that of the British Isles not unfavourable, 
it seems reasonable to presume that the greater floristic diversity of 
Jamaica is due to its proximity to the Central American region of 
continuous evolutioTTtYorrFearTy geological times. Ceylon, on the 
other hand, although also tropical and humid and likely to be capable 
of supporting a large flora, is near the geologically much_ younger 
and floristically poor er so uthern portions of India, and consequently 
has hadT~aTess^diverse flora to__draw_-upon. In the British Isles, the 
flora and proportion and also degree of endemism are all smaller 
than in Jamaica and Ceylon, owing in part to relatively recent 
glaciation, in part to the prevailingly less favourable climate, and 
in part to the proximity of (and only recent separation from) the 
mainland of Europe. Greenland, most of whose still greater area 
is covered by ice, has an even smaller flora, owing, it seems clear, 
to the widely inimical climate and severe Pleistocene glaciation. Its 
about 475 species of vascular plants (including well-established 
introductions and 3 1 Pteridophytes) show an extremely low propor- 
tion and degree of endemism, with none at all among the Pterido- 
phytes. (These are the latest statistics which have become available 
since the preparation of Fig. 45, B.) 

6] foundations of modern distributions 173 

Postglacial 1 Changes 

Climatic change has affected plant distribution right up to the 
present time, and indeed such changes and their effects on plant 
life are still going on and presumably will always continue. The 
postglacial sequence of changes is best known and perhaps most 
marked in temperate Europe, where the sequence has been briefly 
as follows : 

(1) the earliest deposits following the last glacial recession show 
evidence of an arctic-subarctic-arctic sequence of vegetation-types 
characterized by Dwarf Birch (Betula nana agg.), shrubby Willows 
(Salix spp.), and Mountain Avens (Dryas octopetala s.l.), 2 developed 
under probably rather dry as well as cold conditions except around 
the middle of the period, which persisted for example in the British 
Isles until about 11,000 years ago. This ' Subarctic' period was 
followed about a millennium later by 

(2) a ' Pre-Boreal ' period of variable but milder climate, char- 
acterized by Scots Pine (Pinus sylvestris), Birch (Betula pubescens s.l.), 
and Elm (Ulmus), with Spruce (Picea) dominant in some eastern 
regions. This was in turn followed by 

(3) a ' Boreal ' period of relatively warm and dry ' continental ' 
climate which towards its end supported mixed hardwood forest — 
particularly of Oak (Quercus) with abundant associated Hazel 
(Corylus) in the temperate belt, and persisting there until probably 
seven or eight thousand years ago. Thereafter followed 

(4) a still warmer but wet ' Atlantic ' period characterized by 
mixed Oak and Lime (Tilia) forest, constituting the so-called 
' climatic optimum ' (i.e. for northwestern Europe) that lasted until 
5,000 or fewer years ago. This was in turn followed by 

(5) the more continental, drier ' Sub-Boreal ', which lasted until 
about 2,500 years ago and in which there occurred a reduction of 
bog growth but an increase of Conifers and the entry of Beech 

1 It is said that we are still not out of the Pleistocene epoch and that, con- 
sequently, we should not speak of the Present or Recent or post-Pleistocene ; 
from our point of view, and although because of variability in different places it 
cannot be satisfactorily defined, the time since the last great glacial recession is 
all postglacial, and seems best so termed (informally, with a small ' p '). While 
suggesting this course, Professor R. F. Flint confirms (voce) that the terms Recent 
and Holocene, which are often used for postglacial time, have also not been 
properly defined. 

2 Also, Professor Gunnar Erdtman informs me (voce), by such ' pioneer ' plants 
as species of Artemisia, Helianthemum, members of the Chenopodiaceae, and 
even Ephedra. 



(Fagus) and Hornbeam (Carpinus). In early historical times the 
climate tended to be cool and moist, yielding 

(6) the ' Sub-Atlantic ' period which was characterized by con- 
siderable bog formation. This appears to have given way during 
the last millennium, and particularly in recent decades, to a warmer 
and drier period. Meanwhile the forests have been gradually 
destroyed by Man. 

So far as has been determined, and indeed as might be expected, 
details of the above changes have varied considerably in different 
areas, so that in some more, but in others fewer, phases have been 
recognized. Nor, according to Dr. H. Godwin (in litt.) y were the 
periods as certain and defined as is commonly believed. In general, 
the tendency has been to find less favourable temperatures to the 
north, and especially in the Arctic, where at best conditions 
approximating those of the present-day Subarctic have prevailed. 
Moreover, different workers have placed different interpretations on 
the ' sub-fossil ' and other evidence available, only the simplest 
generalization being applicable to the majority of the drastically 
affected parts of the world. However, there has clearly been a 
climatic succession consisting of three main phases, viz. (i) increasing 
temperature, (2) culmination of warmth-loving trees, and (3) decrease 
of warmth-loving trees and appearance of those predominating today. 
Thus in temperate America there are scarcely any profiles known 
to yield such tundra plants as occurred in the Subarctic of Europe, 
but the other periods are closely comparable with the European, 
comprising a cool Pre-Boreal (Spruce — Fir), a warm and dry Boreal 
(Pine), a warm and moist Atlantic (Oak — Hemlock — Beech, etc.), a 
warm and dry Sub-Boreal (Oak — Hickory), and a cool and moist 
Sub-Atlantic (Spruce — Hemlock, Oak — Beech, etc.). Age-checks by 
such methods as ' carbon- 14 ' have recently indicated that these 
climatic changes on the two sides of the Atlantic approximately 
coincided in time. 

This hypothesis of climatic change suggests that, of the plants 
which may be segregated into groups because of their preponderance 
during one or another of these periods, the Atlantic and Sub- 
Atlantic species favour regions having an oceanic climate and the 
plants of the earliest or Subarctic period do not avoid coastal regions 
nowadays, whereas the Boreal and Sub-Boreal ones favour inland 
areas of continental climate that tend to be drier and warmer though 
often given to extremes. Sometimes the species of one category 
will be so grouped together in an area as to suggest a particular 


postglacial history or origin, such considerations again being 
important as a basis of present-day distribution. So here again we 
see instances of the significance of historical causes for an under- 
standing of the present distribution of species and of their groupings 
in vegetation. 

The Genetical Heritage 

The remaining aspect to be considered as part of the historical 
background of plant distributions is the g enetico-evolutionary j one, 
an outstanding exam ple b eing a ffordej djjypolyploids, treated in the 
next section. It seems reasonable to suppose that trie physiological 
tendencies and habitat preferences of particular plant entities have 
long been much as they are today, as have, doubtless, many of the 
habitats themselves, and that morphological (that is, of general 
form) and anatomical (of internal structure) indications of ecological 
relationships that hold nowadays are also largely applicable to plants 
which lived in earlier ages. Indeed such assumptions are behind 
many of our contentions regarding climatic and other changes in 
bygone ages. Nevertheless, evolution has doubtless proceeded at 
various speeds and with varying results throughout the period in 
which there have been advanced forms of life on earth, and among 
the characters affected have surely been such ones as migrational 
abilities, acclimatization potentialities, and habitat preferences. 
Consequently, we should consider in broad outline the evolutionary 
tendencies that manifest themselves in these characters, and some 
facts and fallacies of resultant areal indication, before proceeding 
with the more practical parts of this treatise. 

Just as most obvious manifestations of form are inherited by each 
generation from the last, and this process is repeated virtually ad 
infinitum, so are different functions, different physiological attributes, 
usually so inherited — including those which control the migrational 
abilities, acclimatization potentialities, and habitat preferences of 
different plants. Indeed these last three groups of factors are best 
considered as one, being in any case all dependent on inherent 
tendencies that can scarcely be separated. An outcome of this 
inheritance, generation after generation, in particular lines or strains 
of plants, is the obvious suitability of certain plants for certain areas, 
the converse holding often more strongly — namely, that certain 
other plants are unfitted for growth in these areas. Just as evolution 
of form results from the action and often interaction of one or more 


processes such as mutation (including chromosomal multiplication 
or other change), hybridization (producing new combinations of 
genie materials), isolation, and natural selection, so does evolution 
regardless of form result in particular physiological make-up. Here 
are included those physiological factors on which depend the suit- 
abilities of particular plants to grow in particular areas. 

By such means is delimited the potential area of a species, outside 
of which it cannot grow naturally. The ultimate limits are accord- 
ingly genetically controlled, being restricted primarily to areas of a 
particular climatic range and secondarily, within these, to areas of 
particular sorts of soils, etc. And even though acclimatization 
allows some latitude, this can only be within genetically fixed limits 
— unless, of course, there is some fundamental evolutionary change 
in the race. This sometimes happens, for example through muta- 
tion or hybridization. Also effective in this direction is some change 
in the population, such as can take place through isolation and 
selection of the forms best adapted to withstand particular local 
conditions. The latter type of instance is not so much genetically 
controlled as genetically allowed, for, as we have already seen, a 
population even of a single species or lower entity usually consists 
of more or less numerous biotypes differing slightly in their inherit- 
ance. Owing to different groupings of biotypes occurring in 
different local populations, or at all events to varying selection, 
different geographical races of a species often appear that have 
different habitat preferences jindj consequently, ranges. And geo- 
graphical isolation promotes evolutionary divergence — not merely 
because of differing selection-pressures and variational tendencies, 
but also because mutations appearing subsequently cannot be shared 
around by interbreeding 4. in time, new species may result. 

In spite of this relative ease and speed of evolution in some 
instances, it appears to be extremely slow in others. Thus it is 
supposed that most at least of the woody species of temperate 
regions date well back into the Tertiary, having existed for five 
million or more years. That, as we have seen, was a time when 
equable climates extended much farther towards the poles than they 
do nowadays, and woody species of temperate regions tended to be 
much more widespread. One of the most active principles tending 
to blur the immediate effects of evolution is introgression, which 
is the gradual infiltration of the germ-plasm (and hence inculcation 
of facets of the character) of one species into that of another as a 
result of hybridization and repeated back-crossing with the original 


parental lines. Disturbance of the habitat is known to favour 
hybridization, and when such disturbance is so common and wide- 
spread as in the Arctic, with the frequent frost-heaving of open 
habitats where there is little competition, we may expect ready gene- 
flow — hence perhaps in part the notorious * plasticity ' of arctic 
plants. For related populations can be lastingly sympatric (that is, 
coexist in the same territory) only if they are reproductively isolated. 
Length of life may be an important factor controlling hybridization 
and introgression, especially where uniformity is maintained by a 
preponderance of vegetative propagation, with consequent restriction 
of gene-flow and limitation of genetical recombination. In such 
circumstances, the ill-adapted offspring tend to be easily eliminated 
by competition. On the other hand, with free gene-flow and a fair 
amount of mutation in rapidly succeeding generations, much new 
1 raw material ' is provided for natural selection to work upon, and 
evolution may be expected to proceed with some dispatch. 

The distribution patterns of organisms, like the external appear- 
ance and genetic constitution of the component individuals them- 
selves, are the end result of the interaction of evolutionary processes 
and climatic, pedological, and other changes over long periods of 
time. Now similarity of distribution patterns suggests similarity of 
general background including evolutionary history, and this may 
give some possibility of divining the history of an organism well 
represented in the fossil record but unknown genetically, provided 
we have others of comparable distribution that are well known 
in this last respect. Such elucidations, and indeed the broader 
ones of plant geography, must, however, be indulged in only 
on the basis of all known facts, and then only with the utmost 

In the light of modern knowledge some old assumptions should 
be discarded or at least greatly modified — for example, that the 
diversity of a group is dependent upon its age, which may be deter- 
mined, at all events relatively, by counting the number of members 
now living, and that the age of a species or other taxon is directly 
related to the size of its area of distribution. Such are the main 
tenets of the hypothesis of ' Age and Area', discussed further on 
pp. 182-3, 20 9- And although ideally there may be some basic truth 
in these assumptions, some aggregate responsible effect in that diver- 
sity may come with time and increase colonization potentialities, even 
as time itself may increase the chances of dispersal, in actual fact 
evolution and migration have proceeded at very different rates in 


different groups and at different periods, both because of inherent 
differences and of being rarely unimpeded. 

Again, such generalizations as those which seek to give directions 
for determining the place of origin of a particular plant group are 
apt to be dangerous, as for example the supposition that the original 
home of a group is the place in which the largest number of its 
representatives exist. Thus old groups have frequently survived 
great alterations of climate and have died out in major regions where 
they formerly flourished — sometimes, with little doubt, including 
those in which they originated — and have apparently found secondarv 
centres of diversity in favourable areas where, it may be expected, 
conditions for evolution are different. And it should be noted that 
for genera and higher taxa of Mammals, where the fossil record is 
far more nearly complete than with plants, the evidence is largely 
contradictory to this hypothesis of diversity indicating the centre 
of origin. Nor is the newer generalization, that the centre of origin 
is the area in which the most advanced species are found, any less 
dangerous — especially with its stated corollary that the most primi- 
tive species will be those remote from this centre. Indeed, in many 
groups of plants, the more advanced members differ from the 
primitive ones in being more effectively specialized for dispersal and 
more genetically ' open ' for migration, so that they may be expected 
to overtake their ancestors in colonizing the earth. 

Finally, even the common assumptions of a single (poar or 
tropical) region of origin and differentiation of the groups of Ihigher 
plants, and of a simple basis for their migration from one continent 
to another, are presumptuous in view of the present meagre state 
of our knowledge and the assertion that the main centres of mam- 
malian differentiation have been at middle latitudes in the interiors 
of the large land-masses. For, as we have seen, evolutionary change 
and migration are fundamental activities that seem to go on practically 
all the time everywhere, though at very different rates in different 
instances and places. 

Polyploids and Their Areas 

The plant geographical implications of polyploids (organisms 
whose body-cell nuclei contain more than two haploid or single 
chromosome sets) have been so much discussed in recent years that, 
although most conclusions still remain tentative, the subject needs 
to be mentioned here. Polyploids are found in most of the major 



groups of plants and, in the Angiosperms, are particularly plentiful 
among perennial herbs. They are often more vigorous than their 
diploid (that is, with body-cells having double the number of 
chromosomes basic to the species or group and characteristic of the 
reproductive cells) relatives even of the same species, and, in addition 
to anatomical differences su ch as larger cells and pollen grains, may 
show morphological deviations such as usually larger flowers and 
coarser stems. Nevertheless it is customary, unless these differences 
of form are striking, to keep related diploids and polyploids in the 
same (major) species. Polyploid s also exhibit a greater tendency 
t o adopt vegetative or asexual means of reprodu cti on than relate d 
diploids. Even more important from our point of view is the fact 
that they may show very different ecological preferences and geo- 
graphical distributions from diploids, though no definite rules can 
safely be formulated to govern their behaviour in this respect. 

It has been widely contended that polyploids are .more hardy 
and conse^uejitly^riore^ior^erly ^ (in the northern hemisphere) and 
high-alpine, in dis tribut io n th anjjiejliploids frorn^ which they have 
been derived. About this there is, however, no unanimity of 
opinion — largely because there are numerous exceptions to what still 
appears to be a distinct tendency. Suggestions that polypl oids ar e 
unusually prevalent in hot and dry regions and that they favou r 
coastal rather than inl and areas, seem to be based on less factual 
e viden ce and, indeed, to be without adequate foundation when the 
situation is viewed on a sufficiently broad basis. There does, how- 
eve?7~appear to be some tendency for polyploids, especially when 
they have arisen through hybridization (allopolyploids), to have a 
wider_geographic range than diploids : thus of ioo examples 
assembled by Professor G. L. Stebbins as recounted by him in his 
book Variation and Evolution in Plants, 60 polyploids were more 
widely and 33 less widely distributed than their diploid relatives. 
It is thought that the proportions of polyploids showing wider 
distributions would be higher if the examples were limited to closely 
related pairs of entities, such as polyploids and their more immediately 
ancestral diploid progenitors. There are also indications that poly- 
ploids_may be more prevalent in regions that were glaci ated in th e. 
Pleistocene than in those which were not, arid in the peripheral 
areas or near the ecological boundaries rather than towards the 
centres of distribution of particular plant groups. This tendency 
evidently goes hand in hand with variation, and results from the 
fact that polyploids have changed reaction norms. As Professor 


S. A. Cain writes in his Foundations of Plant Geography : ' Ecological 
advantages may arise from the competitive ability of the polyploids 
that allows them to associate favorably with or even to replace their 
progenitors, or from the capacity of the polyploids to occupy new 
climatic or edaphic situations, and hence areas in which they are 
not confronted with competition from their close relatives.' 

Altogether it seems that the phenomenon of polyploidy may have 
considerable significance in ecological and geographical connections. 
Thus some of the changes which are apt to accompany polyploidy, 
such as alterations in plant stature and leaf-size, in the frequency 
and size of stomata, and in hairiness, may affect transpiration and 
hence the water economy of the plant. Although some of these 
characteristics are evidently beneficial to polyploids, others, such as 
their commonly observed retarded rate of development and lateness 
of flowering, may militate against their own good, weakening their 
competitive ability. Various physiological changes have also been 
observed to be associated with polyploidy, including changed cold- 
and perhaps drought-resistance which may have great survival and 
hence phytogeographical significance. The outcome, however, 
apparently varies with circumstances and with the particular case 
under consideration. The same is true of life-form changes, 
perennials being often polyploid in contrast with their annual 
relatives. These and other features affect the adaptability and com- 
petitive power of the plant and hence its ecological amplitude and, 
consequently, geographical distribution. Moreover, any tendency 
it may have to dominate is thereby affected, and, where dominance 
is concerned, so is the habitat and, ultimately, the distribution of 
other species. In view of the ease with which polyploidy may now 
be induced in plants by various laboratory practices, it may be that 
this tendency will become of even greater importance in the future 
than it is today — for example in the production of larger and better 
crop plants. On the other hand, it should be recalled that some 
species which lack the benefit of chromosomal races ' do ' just as 
well as those with polyploids, showing great ecological and geo- 
graphical amplitude owing probably to a richness in biotypes or 
larger genetic entities. 

Further Consideration 

E. V. Wulff. An Introduction to Historical Plant Geography (Chronica 
Botanica, Waltham, Mass., pp. xv -\- 223, 1943) ; for various 
palaeobotanical and allied aspects. 


W. C. Darrah. Principles of Paleobotany (Chronica Botanica, Leiden, 

Holland, pp. [vi +] 239, 1939). 
R. F. Flint. Glacial Geology and the Pleistocene Epoch (Wiley, New 

York, pp. xviii 4- 589 & maps, 1947). 
R. F. Flint. Glacial and Pleistocene Geology (Wiley, New York, pp. 

xiii -j- 553 and 5 additional maps, 1957). 
S. A. Cain. Foundations of Plant Geography (Harper, New York & 

London, pp. xiv -f 556, 1944) ; for a philosophical discussion of 

the origin and history of plant types and areas. 
G. L. Stebbins. Variation and Evolution in Plants (Columbia University 

Press, New York, pp. xx + 643, 1950). 
Nicholas Polunin. Arctic Botany, vol. I : Exploration, Taxonomy, 

Phytogeography (Oxford University Press, London etc., in press) ; 

for application to the northern regions of the world. 
H. Godwin. The History of the British Flora (Cambridge University 

Press, Cambridge, Eng., pp. viii -f 384 and additional table, 1956) ; 

for application of recent stages to a more limited area. 

An interesting instance of protracted persistence after introduction was 
noted by the present writer in 1936 in southwestern Greenland, where he 
discovered living descendants of plants which had evidently been intro- 
duced from North America by the Norsemen whose Greenland settle- 
ments are known to have died out several centuries ago. As certain of 
these plants are of known but restricted (in two instances barely over- 
lapping) distribution on the eastern North American seaboard, they give 
a clear indication of where their ancestors probably came from, and where, 
accordingly, Viking relics should be sought which would prove once and 
for all that North America was known to Europeans long before the birth 
of Columbus. 

In connection with the wide acceptance of sub-fossil pollen grains as 
evidence of former climates, the author cannot forget that through much 
of the summer of 1950 he found the most plentiful pollen in the air near 
the ground in West Spitsbergen to be that of Pinus sylvestris, the nearest 
trees of which were growing on the Scandinavian mainland several 
hundreds of miles away to the south. This indicates the need for caution 
in interpretation — including the desirability of statistical comparisons 
and, above all, avoidance of any tacit assumption that a small deposit or 
reasonable amount of an airborne pollen was necessarily produced locally. 

Chapter VII 


In the last chapter we considered what lies behind the geographical 
distributions of plants as we see them today. We must now concern 
ourselves with those distributions that appear to be natural, leaving 
until the next chapter the ' artificial ' ones which have obviously 
been made or modified by introduction or other interference by Man. 

Each different kind of plant has its own particular distribution or 
range, which is dependent, as we have seen, on its history, migrational 
ability, and adaptability. Indeed, it is doubtful whether any two 
of the hundreds of thousands of different kinds of plants known to 
science have precisely the same distribution ; and in any case 
distributions are changing all the time. It is consequently impractic- 
able, and wellnigh impossible, to consider such matters in detail ; 
yet when a broad view is taken many interesting facts stand out, and 
generalization may be valuable. For whereas any hard and fast 
system of classifying ranges would be artificial, because it would not 
reflect the natural diversity observed, some useful categories can, 
and for practical purposes should, be widely recognized. 

The term area (or range) in plant geography is applied to the 
entire region of occurrence of a particular taxonomic entity (taxon, 
plural taxa) or vegetational unit (econ, plural eca). Within this 
range it is often necessary to consider the local distribution, some- 
times called ' topography ', which at best is no more nearly continuous 
than are suitable habitats for the entity or unit in question. For 
whereas climatic limits usually constitute the chief boundaries of 
plants, local topographic, edaphic, and biotic factors are all apt to 
have their effect — as will be explained and illustrated further in 
Chapter X. This, albeit secondary, effect is usually considerable, 
often drastically limiting the areas of plants, within the bounds 
prescribed by climate, to those offering otherwise favourable 

Mention should be made here of the hvpothesis of ' Age and 



^rea!^ which claims that the area occu pied by a species is propor- 
tionate to its age (i.e. the time it has existed). In spite of what 
has just been said, this is often true and is indeed somewhat axiomatic, 
as reconsideration of dispersal and migrational ' mechanics ' would 
lead us to expect. For iftwo or more species with identical capacities 
in these respects begin their migrations at different times, the earliest 
starter will be found, at ajvy_p^rticulj^time^to have extended the 
farthes t. Yet actually there are so many superimposed factors 
causing complications, and such numerous and often striking excep- 
tions, that the hypothesis is of very doubtful value, and at best may 
be considered a mere generalized working one — cf. p. 209. 

Before proceeding to the main topics to be considered in this 
chapter, we should explain the additional taxonomic concepts of 
ecads, ecotypes, and clines. An ecad as understood in this work 
is a plant type or form produced within the life-time of the individual 
in response to a particular habitat factor. An ecotype is a distinct 
race resulting from the impress or selective action of a particular 
environment. A dine is a geographical or ecological gradient in 
phenotypic characters {i.e. physical make-up). These entities are 
below or outside the usual specific bounds but should be borne in 
mind as exhibiting much the same geographical characteristics as 
do the usually higher taxa which we are more prone to consider. 

' Continuous ' Intercontinental Ranges 

Except perhaps if it is very limited, the area of a taxon or of a 
vegetational feature is never really continuous ; in reality all manner 
of interruptions occur, resulting in some characteristic topography 
(using this term as implying local distribution-pattern). Neverthe- 
less we tend to refer to those distributions which involve spreading 
over a whole territory as ' continuous ', at least provided the various 
stations are not more widely separated than the normal dispersal- 
capacity of the plants concerned. 

Among the most frequent causes of interruption is the lack of 
suitable habitats, which indeed may themselves be widely separated 
or sparsely distributed. In such circumstances it is a matter of 
proportion, and consequently of opinion, as to whether a particular 
range should be considered continuous or otherwise. Thus whereas 
the Sea-beach Sandwort (Arenaria peploides agg.) is found on almost 
all sea-shores of temperate and boreal regions, where its distribution 
may in the wide sense be considered virtually continuous, it is 


usually absent inland, and accordingly in the floras of many individual 
regions it is either lacking or actually of disjunct distribution. 
Again, a ' continuous ' area may have ribbon-like prolongations 
extending beyond its main boundaries and even lack continuity in 
these prolongations, especially when they are narrow, as for example 
along river valleys which are interrupted by narrow gorges. 

Of continuous intercontinental ranges we may consider four main 
types : the cosmopolitan, the circumpolar, the circumboreal (or, 
alternatively, circumaustral), and the pantropic. 

(1) Cosmopolitan — distributed all .over the globe. In reality no 
species is truly so, or, probably, found in all edaphically similar and 
hence potentially suitable habitats. Thus even without the funda- 
mental effect of climate and the common interference of other living 
organisms, it seems unlikely that any one kind of plant can be reallv 
cosmopolitan. Those which most closely approach being so are 
the ones which are least exacting in their habitat requirements, 
tending to be ubiquitous. These wide-ranging plants which tend 
to be indifferent to environmental conditions are called ' cosmo- 
polites ' or ' pan-endemics ' ; in view of the merely relative nature 
of the condition, the newer term ' semi-cosmopolite ' seems more 
accurately descriptive of them. They should at least occur in all 
of the six widely inhabited continents. Actually, outside of weeds 
of cultivation that have followed Man, they seem to be confined to 
the lower groups of cryptogams. 

(2) Circumpolar — distributed around the North or South Pole. 
This term, again, has been used far too commonly and loosely. It 
seems desirable to apply it only to plants which reach the arctic or 
antarctic ' polar ' regions, wherever else they may occur, and pre- 
ferable, at least in the present writer's opinion, to accept as ' arctic 
circumpolar ' only those plants which occur at least in all of the 
ten sectors into which he has divided the Arctic for such purposes. 1 
For these plants are truly ranged around the North Pole. Whether 
such criteria can be used in the case of the Antarctic has not yet 
become clear. Even if the limits of the Arctic are rather narrowly 
set, so as to exclude for example the whole of continental Scandinavia 
and Iceland, there are rather numerous arctic circumpolar species 
already known among the higher plants, and many more among the 
lower cryptogams which tend to be relatively easily dispersed and 
less exacting in their habitat requirements. Still others will, clearly, 

1 Cf. Circumpolar Arctic Flora (Clarendon Press, Oxford, pp. xxviii - 514, 


Fig. 46. — Map showing arctic circumpolar distribution as exemplified by Edwards's 
Eutrema (Eutrema edwardsii). (From data kindly supplied by E. Hulten.) The 
broken line indicates the southern boundary of the Arctic as proposed by the 
present author, and the 10 sectors (given Roman numerals I-X) into which 
the region is divided are those he employs in phytogeographical citations. 

1 86 



be added with further exploration ; meanwhile familiar examples 
of flowering plants belonging to this category are the Purple Saxifrage 
(Saxifraga oppositifolia agg.) and Edwards's Eutrema (Eutrema 
edwardsii {see Fig. 46). 










f X!yp- . 


( / * 

Cky Js >].-.■ v. 





y \< 



% J> 




Fig. 47.— Maps showing circumboreal and circumaustral distributions. A, Kibes 
spp. (Currants and Gooseberries), circumboreal (after Hutchinson), but omitting 
some arctic stations; B, southern species of Danthonia (Poverty-grasses and Wild 
Oat-grasses), circumaustral (after Fernald) ; also northern range, but omitting a 
station in southwestern Greenland. 

(3) Circumboreal (or circumaustral) distributed around the top 
(or bottom) of the world in the boreal (or austral 1 ) zone. It seems 
desirable to separate this category from the circumpolar, though 
clearly a plant can belong to both, as in the case of the Purple 

1 Beware confusion with the other uses of this word in biogeography. 


Saxifrage, which is also alpine (cf. Fig. 49). Indeed, most circum- 
polar plants are at the same time circumboreal, though the converse 
is_by no mea ns true. The bor^Tand"austral zones lie next to the 
arctic and antarctic ones, and seem best considered as extending to 
the border of the subtropics. Examples of groups having such 
distributions are shown in Fig. 47, the circumboreal being the genus 
Ribes (Currants and Gooseberries) and the circumaustral the southern 
species of Danthonia (Poverty-grasses and Wild Oat-grasses). 

Fig. 48. — Map showing pantropic distribution of the Palm family (Palmae). 

(After Good.) 

{j j Pantropic — e xt ending p ractically throughout the tropics and 
subtropicj^ Jeastjwidespread in the tropical regions~oT^sia, 
Africa, and .America. A fine example is the Palm family (Palmae), 
as indicated in Fig. 48. Most, but by no means all, pantropic 
species appear to have been introduced by Man through much of 
their range. 

It may be noted in the above that when a very wide view is main- 
tained, mere outliers can be overlooked and even oceans practically 
ignored, distributions across them being considered continuous. 
Moreover, as we proceed from the poles to the tropics and the 
distances involved expand, there is a tendency for fewer and fewer 
minor taxa to be circumglobal. Indeed, in the tropics, subtropics, 
and adjacent warm-temperate regions, it is not uncommon for whole 
genera and even larger groups to be limited to closely adjacent or 
even single land-masses. 


Discontinuous Ranges 

In discontinuous or disjunct ranges the plants are separated by 
wider gaps than the dispersal capacity of their propagules would 
normally bridge. Sometimes the distinction from the so-called 
continuous ranges is doubtful, in being a mere matter of degree, but 
often a taxon will inhabit two or more widely separated areas whose 
elucidation may be a difficult matter. In many cases, however, 
areas once thought to be entirely distinct have been found to be 
otherwise on exploration of intervening tracts, in which the plant 
or plants in question have appeared, and consequently supposed 
gaps have been closed (cf. Fig. 62). In yet other cases, ranges may 
if desired be considered continuous provided there is an absence 
of any suitable habitat between the colonized areas, though here 
again a discreet sense of proportion must be exercised and such 
major barriers as oceans and ice-caps often recognized. 

The above discussion affords instances emphasizing the need for 
caution in describing the distribution of a particular plant — especially 
if it is of a small or insignificant nature, or is adapted to a limited 
range of habitat conditions and consequently has a ' fragmented ' 
and complex topography. But provided these warnings are borne 
in mind and no inflated body of theorizing is based on unwarranted 
supposition, the concept of discontinuity of area is a very real one 
and has greatly stimulated research and philosophical speculation 
in plant geographical and allied fields. 

As for the main causes of discontinuity (apart from the controversial 
extremes of sudden long-distance dispersal or ' historical ' wiping 
out in intermediate areas, both of which have obviously taken place 
in the past), they are usually environmental in being due to particular 
topographic, climatic, edaphic, or biotic characteristics which lead 
to areas being separated from each other by tracts of different 
character. This sets aside for the time being the possibilitv of 
polytopic origin [see pp. 206 et seq.). 

We should mention some of the general Jypes of .disc ontinuity ? 
regardless of cause. An area is described as (1) diffuse when jt is 
hroken up into small^jnor e or less numerous and equ al _parts ; 
(2) bipartite when it is composed of only two separate parts in the 
same hemisphere, one of which is extensive and forms the main 
part and the other of which is subordinate ; (3) bipolar when it is 
composed of two parts widely separated in the northern and southern 
hemispheres ; (4) altitudinal when it is composed of one part 


\ ->\"~A— 

/ ~-~- 

/ ^\/[ P 



s/l X\pS 

^%ri! }\ 

*W )>C \ 

^\/i - — " 

Sc?""""«^ X&Zo&J- 

* 1 _>» M J ~~ " — — / 

\\/ +L^ i \ 

? 7i 


Mr i/ & \ •) ] 


^~---.^Li \ 

3 \\**/> 

^^--7 •••/. 

^f OH? 

t\ \ ^1 

/ /\ '' 

^u? •-Ay'v, 

/\ /' 

/\ x^vL )S^\ 

■^— >^fJ / • Y 

V^^ V' x ' 


/ ^ )W^^^dL___i 

\)\"i , v . 

Fig. 49. — Map showing arctic-alpine distribution as exemplified by Saxifraga 
oppositifolia agg. (from data kindly supplied by E. Hulten). 



Fig. 50. — Map showing range of Drooping Ladies'-tresses {Spiranthes rotnan- 

zjffiana) (North Atlantic, etc., distribution). (After Fernald, emended according 

to directions given by E. Hulten.) 

Fig. 51. — Map showing range of Skunk-cabbage (Symplocarpus foetidus) (North 
Pacific distribution). (After Fernald.) 

FlG. 52. Map showing North-South American distribution of Pitcher-plant 
family (Sarraceniaceae). (After Hutchinson, emended.) 


^J^" : ' 


/ «a 

>\ ^5^^ >» 

'\^<x\' c 

' //b\ / " N / ^ x\\ 

VV^ ^S^Ir^^T' 


/ /)Vyj\ 

/ / • - M ^^ i<^<; ^A ^^^ \\ 

W^c\ \f 



■Ok •( 



\ \ \ \r i* ^^V^ 

^\ 1 / ^"^Ac \ / ' 

\. \ / <^//^^ \ / ' 

\c! / ,s / ( ^^\V ' 

&/ J 

7\ **( / /J /V-^ ' 


^Lai— — ^r ~lv \ 


>^X \ 

^^l \ 


'"' \ \ 

Fig. 53. — Map showing range of Cimicifuga foetida (Europe-Asian distribution). 
(Contributed by E. Hulten.) 



situated inj^ne altitudinal zone and another in another L_zone not 
directly adjoinin g ; and (5) when diffuse, bipartite, or otherwise, 
and populated by identical forms, it is said to be homogeneous — as 
opposed to the heterogeneous discontinuity that involves related or 
vicarious forms occupying different component parts of the range. 

As for more specific types of discontinuous ranges, we may 
mention the following as being among the most familiar and 
important : 

(1) Arctic- Alpine — distributed in the arctic region and in moun- 
tain systems of temperate or even warmer zones ; examples are the 
Herb-like Willow (Salix herbacea) and the Purple Saxifrage (Saxifraga 
oppositifolia agg.), see Fig. 49. 

Fig. 54.— Map showing range of species of Platanus (Plane-trees, or Buttonwoods). 
(After Fernald, emended.) 

(2) North Atlantic — distributed in North America and Europe, 
and sometimes also locally in Asia ; examples are the familiar 
Bog Club-moss (Lycopodium inundatum) and the Hooded or Droop- 
ing Ladies'-tresses (Spiranthes romanzoffiana), see Fig. 50. 

(3) North Pacific— distributed chiefly in North America and 
Eastern Asia, though sometimes elsewhere ; examples are afforded 
by different species of Torrey Pine (Torreya) and by the Skunk- 
cabbage (Symplocarpus foetidus), which is one of that remarkable 
group of species common to eastern Asia and eastern North America 
but wanting in the regions lying between — see Fig. 51. 

(4) North-South American — distributed in North and South 
America but lacking continuity between ; an example is afforded 
by the members of the Pitcher-plant family (Sarraceniaceae), see 

Fig. 52. 

(5) Europe-Asian — distributed in Europe and Asia but lacking 
continuity between ; examples are Leontice altaica and Cimicifuga 
foetida, see Fig. 53. 


i 9 4 


(6) Mediterranean — various types including the European and 
African shores of the Mediterranean Sea, or the Mediterranean 
basin and some distant continent — as in the species of Platanus 
(Plane-trees, or Buttonwoods), indicated in Fig. 54. 


' VI 






s (F 

Fig. 56. — Map showing range of the genus Jovellana (Scrophulariaceae) (South 
Pacific distribution). (After Hutchinson.) 

Fig. 57. — Map showing range of the genus Asclepias (Milkweeds, or Silkweeds) 
(South Atlantic, etc., distribution). (After Good.) 

(7) Tropical — distributed in two or more separate tropical regions 
such as occur within the Old World (palaeotropical discontinuity) 
or the New World (neotropical discontinuity) or both (pantropical 
discontinuity). Among the various minor types, two are illustrated 
in Fig. 55, one being of a genus and the other of a family. 

(8) South Pacific — distributed at least in South America and New 
Zealand, as in the case of the genus Jovellana illustrated in Fig. 56, 
and often also in other Pacific islands and in Australia. 



(9) South Atlantic — distributed at least in South America and 
Africa (often including Madagascar), as in the case of rather numerous 
genera including Asclepias (Milkweeds, or Silkweeds), see Fig. 57. 

(10) Antarctic — distributed on the antarctic mainland (usually 
as fossils) and in the southern parts of South America, New Zealand, 
or on some other austral island or islands, as in the case of the 
genus Nothofagus (the so-called ' Beeches ' of the southern hemi- 
sphere), see Fig. 58. 

These ten major types do not by any means cover the detailed 
diversity of discontinuous areas even of an intercontinental nature. 

Fig. 59. 

-Map showing (bipolar) range of the genus Empetrum (Crowberries). 
(After Good, emended.) 

Thus in the general category there is the ' bipolar ' type mentioned 
earlier, an example of which is afforded by the genus Empetrum— 
see Fig. 59. Other well-known examples are afforded by certain 
Sedges. Also involving parts of more than one continent is the 
' Gondwana ' type which, in conformity with the palaeogeographical 
basis of ' Gondwanaland ', tends to embrace Africa, Madagascar, 
India, and Australia. 

Even more numerous than the above are the tvpes of discontinuous 
range of an m/racontinental nature, as for example in Australia 
where the section Erythrorhiza of the genus Drosera (Sundews) 
affords a striking example as indicated in Fig. 60, or in southwestern 
Kurope with the ' Lusitanian ' clement in the British flora. This 
last element is composed ideally of plants which grow in oceanic 


western Ireland and only reappear, at least so far as natural occur- 
rence is concerned, in some such distant region as the Iberian 
Peninsula, an example being afforded by Mackay's Heath {Erica 




Fig. 61. — Map showing 
' Lusitanian ' distribution 
of Mackay's Heath (Erica 

Fig. 60. — Map showing intracontinental discontinuous dis- 
tribution in Australia of the section Erythrorhiza of the 
genus Drosera (Sundews). (After Diels.) 

mackaiana), see Fig. 61. It has been suggested that the northern 
stations in such instances may be relics of a postglacial warm period 
or, in the case of some still more widely disrupted ranges involving 
the tropics, of the interglacial ' pluvials \ 

Relic Areas 

These, as supposedly in the cases just mentioned, are areas 
occupied by relic species (often called ' relicts ') which in the 
phytogeographical sense are remnants of an earlier flora that have 
been ' left behind ' when surrounding areas have been vacated. 
Thus relic areas themselves normally constitute remnants of once- 
extensive areas, being usually isolated and often contracting. 

The determination of whether a species is really a relic is not 
always an easy and definitive matter, and the same consequently 
applies to relic areas. Even when the usual criteria are applied we 
may go astray, as in the case of some of the plants whose apparently 
disrupted ranges were formerly supposed to indicate that they had 
persisted on unglaciated ' nunataks ' through the Pleistocene glacia- 
tions; yet when intermediate areas were properly explored the 

FlG. 62. — Maps showing known localities of Low Sandwort (Arenaria humifusa). 
A, as published by Nordhagen (1935); B, as published by Polunin (1943); C, 
as published by Porsild (1957) for the New World only, including areas that have 
been heavily glaciated and islands that have only recently emerged from the sea. 
Still further localities are now known. 



plants were found there too, the ranges being in fact almost as nearly 
continuous as habitat suitability allowed. Fig. 62 illustrates this 
point. Ideally the finding of fossil remains of the plant in question, 
in surrounding areas where it does not now grow, will demonstrate 
its relic nature and indicate from what period of time its local 

Fig. 63. — Map showing recent (dots) and ' fossil ' (circles) stations of the Water- 
chestnut {Trapa natans) in Scandinavia. (After Hulten.) 

occupation dates. This is illustrated by the present-day and ' fossil ' 
stations of the Water-chestnut (Trapa natans) in Scandinavia 
(shown in Fig. 63) and of the giant Redwoods in the northern 
hemisphere in general (indicated in Fig. 43). In this connection 
it is interesting to note that whereas Trapa is now known in North 
America only as an introduced weed that is apt to be aggressive, 


there is much fossil evidence that it was formerly widespread there 
in the wild state. 

In general the main, non-relic part of an area either occupies its 
original territory, any outlying parts having become separated owing 
to habitat changes, or results from secondary dispersal, in which 
case the relic part may actually lie within the main area. An example 
of this latter type is afforded by the Scots Pine (Pinus sylvestris), 
which has relic portions of its area in the mountains of Europe, 
whereas its extensive occupation of surrounding sandy lowlands is 
the result of secondary colonization. Such colonization may even 
take place from a relic area, for example when climatic conditions, 
whose deterioration in surrounding areas had led to isolation, 
subsequently improve so that suitable plants can recolonize around 
the isolated area. Much the same often occurs with the removal 
of heavy grazing which had caused the restriction of some plants 
to limited areas (that in a sense were then relic ones), whence thev 
spread again when the animals moved off or died. 

A species which occupies a relic area throughout its range may 
be termed an ' absolute relic ', while one of which only an isolated 
part of the area is relic is known as a ' local relic '. An 'endemic 
relic ' is one that is restricted to a single region. Those relics which 
achieve a secondary distribution by the occupation of suitable 
habitats are known as 'migrant relics', the more recently occupied 
habitats being termed ' pseudo-relic ' and the plants ' pseudo-relics ' 
(or ' -relicts '), though these terms may also be applied in cases where 
a plant acquires the apparent character of a relic without actuallv 
being one. Most of these principles may be applied equally to 
single species or other taxa, groups of species (as in relic colonies), 
or entire floras. 

Many relics, having long been in partial disharmony with present- 
day habitat conditions, have become depauperated in biotvpes ; the 
consequent loss of the capacity for variation and adaptation has led 
to their being considered conservative. Such plants have often been 
called ' senescent '. Being commonly restricted to narrowly specific 
environmental conditions, they may fail to retain even their limited 
area, in extreme cases becoming extinct. However, if favourable 
conditions should reappear and, for example, allow the approach of 
formerly segregated variants so that hybridization can take place and 
new forms thereby arise, the species may take on its former vigour 
or even exceed it and become aggressive. 

Leaving aside the so-called ' anthropogenic relics ' whose areas 


have become drastically reduced through the activities of Man, and 
the ' cultivated relics ' whose sown area, owing to their low economic 
value, has been reduced to small compass and few localities, we 
may usefully distinguish three main classes of relics on the basis 
of the type of natural habitat-change which has isolated them. 

(i) Formation relics that occupy limited areas within the boundaries 
of major plant communities (formations) which have undergone 
considerable changes in composition. Striking examples are the 
residual wooded tracts that are sometimes found in some extensive 

(2) Geomorphological relics that are connected in their habitat 
preferences with particular ecological conditions but that, owing to 
edaphic or allied changes, are no longer provided with the conditions 
of growth to which they are accustomed. Familiar examples include 
marine plants inhabiting freshwater lakes, and shore plants growing 
along the edges of dried-up gulfs. 

(3) Climatic relics that give evidence of having originated and 
formerly flourished under other climatic conditions than those in 
which they now grow. Examples are the mesothermic plants to 
be found in some boreal areas that have cooled at least since the 
' postglacial optimum ' when such plants presumably migrated to 
these areas. 

While it is scarcely possible to distinguish further classes of 
' biotic relics ' resultant on grazing, etc., as distinct from those 
engendered by Man or his animals, or on plant competition with 
our present limited knowledge thereof, there is another basis on 
which relics may be classified, namely, their age and origin, the 
main such classes being : (1) pre-Tertiary, (2) Tertiary, (3) glacial, 
(4) interglacial, and (5) postglacial relics. 

Vicarious Areas 

These are areas belonging to closely-related taxa (vicariads) 
derived from the same common ancestor and tending to be mutually 
exclusive of one another in naturally {i.e. without human interference) 
occupying separate areas. Sometimes, and especially when their 
ecological requirements differ only very slightly, vicariads may be 
mutually exclusive through being closely competitive. This is the 
case with many subspecies and closely-related species ; on the other 
hand, with higher groupings — and even families and whole com- 
munities may in a sense be vicarious — there is less reason to suppose 


that their mutual exclusiveness is due to competition. From their 
very nature, ecotypes often tend to be vicarious, as do the extreme 
' ends ' of clines. In any event, the process of genesis of geographical 
races seems to be the main basis of the formation of ' vicarious 
areas ', at least among the lower taxa, and this and any subsequent 
segregation tends to take place towards the periphery of the range 
of the ' parent ' species, where the latter is in general least happily 
adapted to the environmental conditions. 

Examples of vicariads are to be found in almost any modern 
taxonomic monograph in which the series of closely-related entities 
inhabiting independent geographical areas are commonly recognized 
as subspecies or even species. These tend to be mutually exclusive, 
although sometimes their areas may show some overlapping ; and 
it is often a matter of opinion (as well as, of course, the degree of 
difference they exhibit) whether they should be termed subspecies 
or raised to the rank of species. This is one of the most persistent 
sources of controversy between the ' lumpers ' and the ' splitters ' 
(of species). Numerous instances are afforded by major species that 
are represented by different minor species or varieties (which most 
often seem best considered as subspecies) on the two sides of the 
North Atlantic — as, for example, the European Royal Fern, Osmunda 
regalis, and its North American subsp. spectabilis [see Fig. 64). 
Polyploids, dealt with towards the end of the preceding chapter, 
are often vicariads (see also p. 204). 

Here it may be well to quote Jordan's ' Law of Geminate Species ' : 
' Given any species in any region, the nearest related species is not 
likely to be found in the same region, nor in a remote region, but 
in a neighbouring district separated from the first by a barrier of 
some sort or at least by a belt of country the breadth of which 
gives the effect of a barrier.' Such pairs of twin or ' geminate ' 
species (or subspecies) actually constitute vicariads, differing in only 
minor characteristics that are of later origin than their common 

True vicariads (which have arisen from a common stock) should 
be distinguished from false ones which have not this close genetic 
relationship. Frequently these last are members of different sec- 
tions of a genus that have developed similar life-forms through 
' convergent evolution '. True vicariads (and consequently the 
areas they demarcate) may be classified according to the manner of 
their separation from one another into (1) horizontal (geographical), 
(2) altitudinal (physiographic), (3) habitat (ecological), and (4) 



seasonal (exhibiting seasonal dimorphism, as in the case of closely 
related forms differing in their times of development). 

Most ' systematic ' vicariads, consisting of pairs or sets of the 
higher taxa which are vicarious, belong to the first or geographical 
category, a good example being afforded by the various races of 
Bracken (Pteridium aquilinum agg.) inhabiting different parts of the 
world. There are also plentiful examples of physiographic vicariads 
inhabiting lower and higher altitudes, respectively — for example the 
Wood and Alpine Forget-me-nots (Myosotis sylvatica and M. 
alpestris agg.), and the Common and Alpine Timothys (Phleum 
pratense and P. alpinum s.L). As examples of ecological vicariads 
growing in different habitats and characterizing different communi- 
ties, we may cite : in fresh and mainly salt marshes, respectively, 
the Bulrush and Glaucous Bulrush (Scirpus lacustris and S. tabernae- 
montanii) ; in soils with high and low available water, respectively, 
the Water and Wood Avens (Geum rivale and G. urbanum) ; and in 
calcium-rich and calcium-poor soils, respectively, the Yellow Moun- 
tain and Brook Saxifrages (Saxifraga aizoides agg. and S. rivularis 
agg.). Such vicariads are commonly intraregional, not infrequently 
growing as closely together as their habitat requirements permit, 
whereas the geographical ones tend to be more widely regional. 

We have already indicated that vicariads, at least of the lower 
taxonomic orders, tend to evolve chiefly about the periphery of a 
migrating ' parent ' taxon. Here new characters are particularly 
prone to arise by mutation or other chromosomal change, and help 
the better-adapted offspring to survive under conditions which are 
less favourable to the parent. Thus autopolyploidy, the pheno- 
menon of multiplication of a plant's own chromosome set, may result 
from extreme habitat conditions and be accompanied by evident 
changes in the plants involved, autopolyploids often having very 
definite geographical ranges differing from those of their ancestors 
which possessed the normal (diploid) number of chromosomes. 
When the changes accompanying polyploidy are very marked, a new 
species may be constituted, which is often a vicarious one. Vicariads 
may also arise as a result of hybridization, which is frequently accom- 
panied by a multiplication of chromosomes (allopolyploidy) ; or they 
may be a consequence of mere local differentiation resultant on 
changes in climatic or other habitat conditions in some part or parts 
of a plant's range. As a result, the initial species may ' break up ' 
into a number of vicarious ones, or of subspecies with distinct 
geographical ranges. 


Endemic Areas 

In contrast with the plants exhibiting various types of discon- 
tinuous range, and which may be widely scattered or at all events 
poly endemic (p oly topic— see next section), are those whose range in 
each case is confined to a single restricted area, not extending beyond 
some one region, island, or other circumscribed tract. Such plants 
are called endemics, although this term again is largely relative. An 
endemic area is the area of a species or other taxon that, in its dis- 
tribution, is limited to some single natural region or habitat, the 
history or conditions of which mark it off from others. Islands and 
mountain massifs are particularly pertinent in this connection. 

Of endemics there are two main types. There are the old ones 
whose range was once far more extensive than it is today, and which, 
being remnants or survivors of former floras, may be called relic 
endemics or epibiotics. These may make up a large proportion of 
the species of ancient islands or mountain massifs, being said to 
involve 72 per cent, of the thousand or so native vascular species 
of New Zealand and 85 per cent, of those of St. Helena. A good 
example of this type of endemism is furnished by the giant Redwoods 
of the western United States, which used to be extremely wide- 
spread in the northern hemisphere (cf. Fig. 43). Their drastic 
contraction in range seems the more remarkable when we recall 
that they include what are probably the oldest individual living 
organisms today, some being reported to exceed 3,000 years in age. 
The other main type of endemic is made up of the relatively young 
taxa, usually below the rank of species, and then usefully termed 
micro- endemics, which are characteristic of newer portions of the 
earth's surface. Thus, when ecological conditions change within 
the limits of some natural region, there is a tendency for new forms 
to evolve, and these may be closely bound to the region owing to 
its special habitat conditions or because they are physically unable to 
spread beyond its confines. Such plants may be called neo-endemics. 

The determination of the proportion of these main types of 
endemism in a particular flora is an important factor in its analysis, 
capable of telling us much about its age and history. Relic 
endemics are particularly useful in indicating antiquity, isolation, 
and diversification of habitats, for these factors all tend to produce 
additional endemics and help in their survival, as probably do also 
suitable conditions for the development of vegetation. Such 
endemics tend to be deficient in biotypes and are usually recognized 


by their relic character and geographical isolation, their small 
amplitude of variation and narrow restriction to particular ecological 
conditions, their relatively small chromosome number, and their 
generally retrogressive nature. On the other hand, neo-endemics, 
being secondarily derived types, commonly have larger chromosome 
numbers ; they also tend often to be relatively aggressive. This is 
especially the case when they are rich in biotypes, for example owing 
to hybridization. 

Apart from the so-called ' pseudo-endemics ' that have been 
encountered only in one place and appear to be mutants, etc., and 
unlikely to persist, there are the ecological endemics which have arisen 
in relation to particular habitat conditions. 

Some endemics are confined to very limited areas, such as a single 
small island or mountain peak, and may be called local endemics. 
Such restriction is usually due (i) to their recent origin (so that 
dispersal has only just begun), or (2) to their antiquity (so that 
the area is a contracted one or even a ' last remnant '), or (3) to 
their high specificity with regard to habitat conditions (which 
prevail only at a given spot within the area that can be reached bv 
viable disseminules), or (4) to the impossibility of expansion (owing 
to physico-geographical obstacles). 

It has been mentioned that isolated islands are often particularly 
rich in endemics ; this is especially the case with those which are 
at least some hundreds of miles from the nearest major land-mass, 
and which may accordingly be termed oceanic. Although the dis- 
tinction is far from definite, it is sometimes useful to think of other 
islands, whose flora bears a closer relation to an adjacent land-mass, 
and which are usually within at most a few hundred miles' distance 
from it, as continental. These islands are phytogeographically like 
fragments of continents or larger islands and are usually inhabited 
by larger numbers of species than are comparable oceanic islands, 
containing as they do both plants and animals for which transoceanic 
transport seems virtually impossible. Some remote islands, such 
as those surrounding Antarctica, are apt to be considered relics of 
a formerly more extensive continent and hence scarcely oceanic. 

Polytopy and the Incidence of Areas 

Polytopy is the occurrence of a species or other taxon in two 
or more separate areas, such species being termed poly topic or 
poly endemic. The discontinuous ranges involving disjunct areas 


which we dealt with earlier in this chapter are examples of polytopy, 
which we must now consider briefly from the point of view of the 
origin of the areas involved. 

Excluding the old conception of special creation, it is yet believed 
by some that approximately polytopic forms may have had an 
independent origin in their existing plurality of areas whose popula- 
tions are similar because of parallel descent from a common ancestor. 
This would explain discontinuity on the basis of the species con- 
cerned having evolved independently in two or more separate areas. 
But whereas this seems possible where a common near-ancestry and 
minor taxa are concerned (for example, through natural selection 
acting on a similar set of mutants), it scarcely seems conceivable 
for members of major species whose common ancestry was remote 
and whose distinctive characters are numerous. Largely separate 
is the contention of the differentiation hypothesis that different 
species of the same genus may have ' crystallized out ' from an 
ancestral complex in two or more areas independently. Some 
students even hold that the polytopic populations in different areas 
have had an independent origin from taxonomically different 
ancestors and have arrived at their present similarity through con- 
vergent evolution. However, in the words of Cain (in the work 
cited at the end of the last chapter) such polyphylesis ' for most 
students of evolution, genetics, and taxonomy, represents only the 
result of inadequate knowledge, or the forming of groups (genera, 
families) for practical convenience, except in cases of hybrid 
descent . . .' 

More widely accepted is the hypothesis that polytopic forms are 
immediately related, the intervening tracts having been ' bridged ' 
in the past either by a continuous series of populations or by long- 
distance dispersal. However, as in other biological instances, it 
seems likely that different explanations apply in different cases. 
Thus concerning dispersal, it has been calculated that even if the 
probability that some member of a population will cross a barrier 
in any one year is virtually nil, during the course of a million years 
the event will be probable, and in ten million years almost certain. 
With regard to the disruption of areas that were previously con- 
tinuous, though not necessarily at one and the same time, we have 
already discussed, especially in the last chapter, such possible causes 
as continental drift, land-bridges, and climatic and other change. 
We have also seen how isolated centres of survival may become 
centres of dispersal, upon the return of more favourable conditions 


— at least if the biotype depauperization of the relic community 
has not been too extreme. 

The centre of origin of an area of other than a lower taxon, at 
least in the absence of extensive palaeobotanical data, is apt to be so 
difficult and hazardous of determination that little will be said about 
it here. For its indication about a dozen criteria are commonly 
employed, though sometimes a fair one may be given by isoflors, 
which are lines delimiting regions supporting equal numbers of 
species, e.g. belonging to a single genus. From the generic centre 
outwards the number of species may be expected to decrease regularly, 
and to assume a pattern which suggests the tracts of past migration, 
which conversely may be followed backwards and found to converge 
upon the generic centre. Even here, palaeobotanical confirmation 
is wellnigh essential. Groups also have their single or multiple 
centres of variation, where there are concentrated the greatest 
diversity and wealth of forms (also called the mass centre), and their 
centres of frequency, where there are accumulated the greatest 
numbers of individuals or stations. 

With single species or lower taxa the situation may be far simpler. 
Thus the tendency to decrease in the number of individuals of a 
species towards the periphery of its area, is closely connected with 
adaptability to definite habitat conditions. For whereas in the 
centre of its area the habitat conditions most nearly approximate to 
the optimum — so that the species can grow under fairly diverse 
conditions, as it often does on different types of soil — nearer the 
periphery of its area anything approaching this optimum is of 
increasingly rare occurrence, and there is often lacking even the 
minimum of conditions required for its normal existence. Thus 
the European Beech (Fagus sylvatica), which ordinarily is capable 
of growing on a variety of soils, is largely confined near the western 
(moist) periphery of its range to the drier calcareous ones. 

As regards an area itself, its shape is best indicated on maps by 
connecting all the peripheral points of its distribution. The shape 
generally depends primarily on the physico-geographical conditions 
of the country, and secondarily on the biological peculiarities of the 
taxon involved. In the frigid and temperate zones the diameter of 
most specific areas is much greater from west to east than it is from 
north to south, whereas in the torrid zone species tend to have a 
relatively larger latitudinal amplitude than elsewhere. 

An area of a species or subspecies usually comes into being largely 
through migration and the operation of barriers thereto, the parent 


species during its dispersal often running into climatic and/or 
edaphic habitat conditions to which it is unaccustomed and which 
in time may lead to modification of the incipient area. From some 
such beginnings further migration usually leads to the current area. 
However, sometimes drastic change may lead to evolution in situ. 
Especially in the cases of young species and of those that have had 
their ranges reduced by relatively recent catastrophes, the area at 
present occupied is only a part, and often only a very small part, 
of what it might be. For, as we shall see in the next chapter, each 
species tends to have, besides its actual area, a potential area which 
may be demonstrated by artificial introduction and is often of great 
practical importance in the regional allocation of crops. It can also 
be of significance with regard to the nuisance caused by weeds. 
Indeed, it seems probable that the majority of present-day ranges 
are by no means complete so far as the occupation of areas of suitable 
climate and soil conditions are concerned. Sometimes this incom- 
pleteness is due, as implied above, either to an insufficient lapse of 
time since the entity evolved, or to the basic inefficiency of its dis- 
persal — or a combination of youth and inefficiency. But probably 
it is more often due to historical changes such as glaciation, or to 
the operation of boundaries set by physical barriers such as seas or 
mountains or deserts, by ecological conditions, or by competition 
with other species. 

Although there is a natural tendency for the areas occupied by 
many plants to increase with age, which can be an important factor 
in biogeographical considerations, the relationship of area to age is 
by no means as direct as has sometimes been supposed. To be 
sure, with some genera and species, especially in certain tropical 
regions, the area of spread is roughly proportional to the age (as 
was suggested by the now unpopular hypothesis of ' Age and Area ' 
— see pp. 182-3), Dut m others this is so far from being the case that 
the area at present occupied gives little or no indication of age. This 
is true, for example, where ancient fossils indicate a much wider 
distribution than now obtains, as in the case of the giant Redwoods 
(cf. Fig. 43). For actually, at least outside of some favoured regions, 
there have been so many, often drastic disturbances that the general 
situation appears to be that the size of an area occupied by a species 
depends less on the age of the species than on other factors. These 
include its adaptability and competition-rigour, the circumstances of 
its genesis, and whether or not ecological conditions and any dispersal 
mechanism or mechanisms have favoured successful migration. 

210 introduction to plant geography [chap. 

Intraneous, Extraneous, and Other Elements 

It is sometimes useful to classify the forms growing in a particular 
territory according to whether in each case the occurrence is well 
within the area of the form (intraneous) or near its periphery 
(extraneous). For instance, the disjunct arctic species occurring in 
the White Mountains of New Hampshire are extraneous there but 
intraneous in their characteristic region of habitation, namely, the 

The components of such groupings form special elements, which 
are severally recognizable in most floras. Thus we may have 
intraneous and extraneous elements in a flora, a preponderance of 
either characterizing certain areas. This leads us naturally to con- 
sideration of specific phytogeographic or floral elements, which are 
closely related to migration. Ideally each such element is the 
floristic expression of a territory of limited extent, in that it involves 
the taxa and phytogeographic groupings characteristic of a given 
phytogeographic area — such as the Mediterranean region, giving rise 
to a ' Mediterranean element '. Often, however, it seems preferable 
to extend this concept of floral elements to include other and much 
wider applications. 

Before a flora can be divided into its main general elements we 
must eliminate all aliens and ' wides ' (also called ' polychores ', i.e. 
species having such an extensive range that they embrace several 
phytogeographic regions). Then the endemics should be studied 
and, as far as possible, assigned to their various categories. There- 
after the remaining species may be divided into groups according 
to the geographical character of their areas, with the object of 
determining the regions whence these groups originated, and so 
perhaps establishing the genesis of the flora. For this grouping, 
five main principles should be followed and elements sought (apart 
from those already accounted for) : 

(i) Geographical elements — grouped according to the types of their 
total areas, their altitudinal ranges, or their distributions within the 
region concerned. This is, however, often insufficient to determine 
the origin of a flora that is not a migration one, for relic and endemic 
elements so grouped do not reflect the genesis of a flora. Even 
with a marked arctic-alpine element it is often doubtful which w T ay 
the components have spread — whether they are arctic types which 
have migrated southward into the mountainous regions offering 
somewhat comparable conditions, or vice versa. All that can be 


done is to divide the species into apparently arctic and apparently 
alpine groups ; and much the same applies to the subarctic or, as 
it is sometimes called, subarctic-mountain element. Other geo- 
graphical elements are fortunately apt to be less vague, examples 
being the Mediterranean and the Atlantic ones. The broader types 
of ecological groupings may also be included here when they char- 
acterize geographical regions — as, for example, certain life-zones and 
formations (or, better, biomes, which are climax formations of plants 
and animals considered together, such as the characteristic Spruce- 
Moose biome of most of the continental regions of Canada). Often 
it is possible to decide on the geographical element to which a 
species most likely belongs by locating its ' mass centre ' (of maximum 
variation), for that is the part of its area which is most likely to be 

(2) Genetic elements — grouped according to their region of origin 
and accordingly reflecting the genesis of the flora. For this, detailed 
monographic study of the groups involved is necessary, and so at 
best it is usually possible to classify in this manner only a few chosen 
species. These first two types of general element are considered 
by some students to be the most important bases for floristic 

(3) Migration elements — grouped according to the routes by which 
they migrated to the region concerned. Examples of migration 
routes are particular mountain passes, river valleys, and suitable 
coasts. Unfortunately, species are apt to reach the domain of a 
given flora by more than one route, so that the establishment of 
particular migration elements is difficult or futile — though often well 
worth attempting, as such elements may provide valuable clues to 
the history of a flora. 

(4) Historical elements — grouped according to the time (such as 
the postglacial climatic optimum period) when they became a part 
of the flora concerned. Further examples are the so-called arctic- 
Tertiary element of evergreen and deciduous trees, and the boreal- 
Tertiary one which included such southerly members as . Palms ; 
these examples represent the inhabitants of the arctic and boreal, 
respectively, regions in Tertiary times, while to their south, stretching 
from southern England to Japan in those far-off days, lay the tropical 
region of megatherms, i.e. plants adapted to high temperatures. 

(5) Ecological elements— grouped according to their immediate 
habitat preferences. Most significant are the oceanic and continental 
elements, embracing, respectively, those species which are adapted 


to a humid maritime climate and those which prefer an arid con- 
tinental one of marked temperature extremes. The oceanic elements 
are generally considered to he the more ancient, the initial land 
flora having been evolved from an aquatic one and further evolution 
having been along lines of emancipation from dependence on water 
and, accordingly, adaptation to more continental habitats. Such 
ecological elements can be of great significance in elucidating the 
history of a particular flora and any major vagaries of climate to 
which it may have been subjected. It should be remembered, 
however, that within the limits of a country or natural region there 
may be found tracts, such as mountain massifs, in which particular 
conditions predominate and which accordingly give refuge to ' alien ' 
plants. These may be termed inclusions, in contradistinction to the 
basic element of types properly belonging to the floral region, and 
the more general penetrants from outside. 

Major Regions 

These will be considered here only in the broadest outline, as 
several subsequent chapters are devoted to them. Vegetational 
regions, being based on life-form rather than on taxonomic proximity, 
may cut across the ranges of systematic units and also differ greatly 
from zoogeographical realms characterized by particular animal 
communities. Fig. 65 indicates the main vegetational-climatic 
regions of the world in highly generalized form, and is the basis 
of the division followed in Chapters XII et seq. It may be noted 
that the western Old World desert region, which is sometimes 
separated as a special one, is here included in the tropical region, as 
the Mediterranean is in the temperate region. This gives us a 
central tropical belt and, to both the north and the south, two others. 
These are the temperate (in the wide sense) and polar belts, the 
former ranging approximately from the polar tree-line and including 
the subarctic and warm-temperate zones, while the tropical belt con- 
veniently includes the subtropics. Each of these broadest of regions 
is itself complex, tending to show latitudinal gradation — so much so 
that accurate detailed maps are scarcely conceivable, at least in our 
present state of frequent ignorance of local features. 

Further Consideration 

Most of the subjects dealt with in this chapter are further discussed 
by Wulff and Cain in their works cited at the end of the preceding chapter. 



For useful examples in several instanees, reference may be made to 
R. Good's The Geography of the Flowering Plants, second edition (Long- 
mans, London etc., pp. xiv j 452, 1953), and for a comparison with 
zoogeographical areas to M. I. Newbigin's Plant and Animal Geography 
(Methuen, London, pp. xv • 298, 1936). 

A valuable example of detailed mapping of the ranges of individual 
species in a well-known region (northwestern Europe) is E. Hulten's 
Atlas over Vaxternas Utbredning i Norden (Stockholm, pp. 119 — 512, 
1950). A somewhat similar project is afoot for the British Isles, and it 
is to be hoped that, in time, more and more regions will be covered in 
this manner. For the ranges as known to date of many different species 
and larger groups, see Die Pflanzenareale (Fischer, Jena, vols. I— V, 

The floras of different areas are dealt with in numerous works to which 
S. F. Blake & A. C. Atwood's Geographical Guide to Floras of the 
World affords the most comprehensive introduction and selective biblio- 
graphy. Part I, covering ' Africa, Australia, North America, South 
America, and islands of the Atlantic, Pacific, and Indian Oceans ', was 
published in 1942 by the United States Government Printing Office, 
Washington, D.C. (U.S. Dept. Agric. Misc. Publ. No. 401, pp. 1-336), 
while Part II, treating western Europe, is to be published by the same 
agency. It is contemplated that a third part will cover the rest of the 

Chapter VIII 


It is now time to deal with the so-called ' artificial ' changes 
wrought by Man, whether intentionally or accidentally, in the 
distribution of plants. In this connection Man seems during recent 
centuries to have been the most potent factor in the world ; and 
as his activity increases and more and more barriers are broken 
down by transport, his effectiveness as a distributor grows ever 
greater. This transportation is quite apart from the changes Man 
brings about incidentally in the course of his ever-extending activities 
of husbandry or desecration. 

From what was said in the last chapter it should be clear that, 
whereas plants have their own distribution patterns, and particular 
taxa have particular areas which they are capable of occupying, it 
is rarely if ever that a vascular plant taxon will occupy anything 
like the whole of the geographical area or areas where the climate 
is suitable for it. Usually, numerous unsuitable habitats will inter- 
vene, and even then there are commonly left areas of suitable 
habitat which the plant in question has failed to reach, or in which, 
if it has arrived, it has failed to establish itself and survive. In 
other words, the present areas occupied by particular plants tend 
to fall far short of the maximum which they are capable of occupy- 
ing : artificial introduction of a plant outside its present natural 
area will frequently demonstrate its ability to grow in a wider range 
of situations both geographically and ecologically. Thus, besides 
its own natural area of distribution, each species has, at least in 
most instances, a wider potential area which, if we include places 
where it can be grown in cultivation or otherwise in the virtual 
absence of competition, is often very much more extensive. This 
principle is of great significance in connection with the introduction 
and production of crops on which Man largely depends. 

The dispersal of plants by Man was considered in a special section 
of Chapter IV. Here we must deal with the results of such transport 
so far as the all-important crops of field and forest and the recognized 



weeds of cultivation are concerned, not forgetting the more devastat- 
ing diseases of those crops. We must also consider the effects of 
cultivation — and of the removal of Man's protection, whether such 
protection had been intentional, for crops, or unintentional, for 

Effects of Cultivation 

It is a common experience of farmers and gardeners that cultivated 
plants and the weeds infesting them or their ground tend to lack 
the ability to spread independently of Man and, in many cases, 
even to maintain themselves unaided. This inability to hold their 
own in the face of natural forces — including, in particular, competi- 
tion from other plants— may even be observed within the natural 
region and habitat range of the more immediate ancestors of the 
cultivated plant or weed. The reason is that Man's influence on 
a plant under cultivation is apt so to change its genetical make-up, 
structure, and physiological capabilities that it is deprived of ' key ' 
advantages in the general struggle for existence. These advantages 
will have been acquired, often by the rigours of natural selection, 
in preceding periods of the plant's evolutionary history, but may 
be lost overnight, as it were, by artificial selection or, more gradually, 
by the ' protection ' afforded by numerous generations of cultivation. 

We may here note a ' round dozen ' of the many and diverse types 
of changes wrought by Man in the make-up and structure of cul- 
tivated (and infesting) plants. 

(i) Genetical changes — involving the loss of characters that are 
obviously beneficial and often needed for the plants' survival in 
natural conditions. Besides the examples indicated among the 
following categories, and plentiful others resulting from such 
activities as artificial hybridization and the induction of polyploidy, 
there are the numerous instances of physiological or otherwise less 
structurally obvious but fundamental hereditary changes. 

(2) Physiological changes — which are commonly hereditary and 
hence included in the above category, but which in other instances 
manifest themselves in the life of a single generation and cause it 
to ' pay the price '. Thus, for example, plants that have not been 
suitably hardened may succumb on transplantation to a less favour- 
able habitat than the one in which they originally developed and 
to which they had become accustomed. 

(3) Structural changes — often bound up with physiological ones, 


and, although commonly hereditary, nevertheless often the result 
of environmental impress during the life of the individual. Seed- 
lings rendered weak by competition with their fellows under other- 
wise favourable conditions of cultivation may not so much as survive 
on transplantation ; or again, thin and delicate ' shade leaves ' of 
some trees are liable to shrivel on exposure. Though drastic, these 
are examples of mere ontogenetic change during a single plant's 

(4) Loss of adaptations for dispersal — or even for the initial act of 
dissemination. The fruits of cultivated Flax and of the Opium 
Poppv do not dehisce when ripe, whereas those of their wild relatives 
do. Again, among weeds, the inflorescences of some noxious 
Brome-grasses break up less effectively than those of their wild 
counterparts, which also tend to have longer awns. 

(5) Loss of protective coverings and sturdiness — for example in 
cereals whose fruits are deprived of the usual outer husks, and in 
the pods of many cultivated members of the Pea family (Leguminosae) 
which lack the fibrous lining characteristic of their wild relatives. 
Presence of the fibrous lining also causes the valves to curl up and 
thus helps dissemination of the seeds. The commonly lesser 
development of fibrous tissue in crop plants is apparently connected 
with their growth in close stands— often protected from winds and 
under conditions of favourable humidity, nutrition, and shading, 
which all tend to promote rapid growth. Similarly, in a dense 
forest the trees usually have tall and slender trunks and weakly 
developed crowns, so that individuals left isolated on removal of 
their neighbours are liable to be blown down, whereas in the open 
the same species tend to be far more sturdy. 

(6) Increase in size of seeds and fruits — usually accompanied by a 
decrease in their number. This tends to reduce their chances of 
dispersal while at the same time reducing the plants' opportunities 
for propagation. Moreover, the production of unnecessarily large 
seeds and fruits is wasteful so far as the plants 5 economy is concerned. 
How much more economical are the fruits of Fireweeds than of 
Pumpkins, and how much more successful as colonists are the 
former plants ! 

(7) Improvement of flavour — of seeds and fruits, which is a com- 
mon objective of cultivation, tends to cause animals to eat them 
more voraciously and completely, and so militates against effective 

(8) Conversion of perennials into annuals — is common in the 


domestication of plants, as for example among the cereals. This 
is advantageous from Man's point of view in speeding up crop- 
production, and may favour the plants' own chances of survival in 
' open ' habitats such as those prepared for cultivation ; but it 
places them at a great disadvantage in competition with natural 
vegetation, which in most undisturbed land habitats is dominated 
by perennials. 

(9) Absence of successful fruiting — for example due to atrophy of 
the sexual organs, to absence of pollinators far from the plants' 
native habitat, or to the desecrations of Man — results in such plants 
being incapable of self-perpetuation by the usual means. 

(10) Seedless fruits — which are often an objective of plant breeders 
for cultivation, likewise render a plant incapable of independent 
existence unless it has some effective means of vegetative propagation, 
in which case it will still lose the benefits of sexual reproduction 
(such as hybrid vigour and the exchange of genie material). 

(11) Double flowers — involving for example the 'conversion' of 
stamens into petals — again render the plant incapable of self- 
perpetuation by the normal means. 

(12) Loss of defensive adaptations — such as spines, thorns, hairiness, 
and hardness — renders the plant defenceless against animal grazing 
and, often, more susceptible to injury from excessive loss of moisture. 

The above features may occur already among wild plants as 
abnormalities, but in cultivated strains they tend to become intensified 
by Man's conscious or unconscious selection and, often, perpetuated 
through his propagation and protection. When no longer cul- 
tivated, or, in the case of weeds, enjoying the benefits of cultivation, 
such horticultural or agricultural strains tend to disappear. Having 
been modified by Man in ways most likely to suit his needs (but 
at the same time harmful to the chances of persistence of the plant 
as an independent organism), they are no longer able to help even 
in maintaining the area of the species to which they belong, at least 
in many cases in the absence of Man's influence. 

The weeds most notably modified through cultivation are those 
that constantly accompany particular crops, thanks to which they 
have long been involuntarily cultivated by Man. Some of them — 
such as, apparently, the cultivated Rye — have become so transformed 
as to be now themselves objects of cultivation. For in weeds, just 
as in intentionally cultivated plants, there tend to be such changes 
as increase in size of seeds at the expense of their number, and loss 
by fruits of their protective coverings and abilities to disseminate. 


A good example of weeds closely associated with particular crops 
is afforded by the so-called ' linicolous ' plants that accompany Flax 
(Linum). These appear to lack the normal adaptations for accom- 
modating their development to seasonal changes, and may even be 
dependent for the completion of their life-cycle upon being gathered 
with the Flax crop when their seeds are ripe, kept in a storehouse 
through the winter, and sown on open soil the following spring. 
In extreme instances the plant has become so modified through 
long association with the crop that its wild progenitors are unknown, 
as is the case also with some crop plants. Examples of such weeds 
of uncertain ancestry infesting cultivated Flax (as indeed their 
specific epithets indicate) are a Campion, Silene linicola, and a 
Dodder, Cuscuta epilinum. There is thus not merely a very close 
association but also a tendency to parallel variation between many 
crops and some of their more commonly accompanying weeds, e.g. 
through their disseminules being difficult, or mechanically impossible, 
to separate from those of the crop itself. 

Naturalization and Acclimatization 

Although, in general, weeds tend to be hardy and to have a very 
wide range of tolerance to differing environmental conditions, so 
that they can spread far and rapidly, at least in ' disturbed ' areas, 
crops are often fastidious in their habitat requirements. Both have 
accompanied Man in his migrations over the world, however, and 
from time to time have given rise to ' escapes ' or, more rarely, have 
become established as naturalized aliens. But it is one thing to 
escape from cultivation or a cultivated area into adjoining terrain, 
perhaps repeatedly and under the beneficial influence of Man, and 
quite a different problem to become sufficiently acclimatized to hold 
sway in a fully wild state in undisturbed habitats among the local 
natives. This latter is a relatively rare feat, as we shall see. Indeed 
for the most part not only crops but also weeds are limited to areas 
that are, or recently have been, in some way disturbed by Man. 

It is sometimes useful when dealing with plants transported out 
of their normal areas to distinguish between naturalization, in which 
they grow under natural conditions that are similar to those to which 
they have been accustomed, and acclimatization, in which they are 
adapted to new environmental conditions differing markedly from 
those of their native habitat or habitats. Although instances of at 
least some degree of the former are common and widespread, there 


are relatively few of the latter unless it be among weeds. For 
acclimatization, except in unusually hardy and tolerant plants, 
involves adaptation to different habitat conditions which is so very 
gradual that the time-span required would be likely to exceed that 
during which Man has been a potent factor in plant distribution. 
It may be expected to involve the natural selection of suitable biotypes 
or more extreme mutants, towards which naturalization is no more 
than a step. 

That such naturalization, at least, is going on widely in the world 
today, is a further indication, if any were needed, that the various 
regions of the globe do not support by any means all of the species 
which could thrive there — at least in the absence of competition. 
However, special studies indicate that, quite apart from the effects 
of competition, plants transferred to regions of seemingly comparable 
habitat may have serious obstacles to overcome before they can be 
considered fully naturalized. These obstacles may be introduced by 
climatic or other environmental conditions which, although thev 
appeared similar, are actually significantly different from those of 
the plants' original habitats (as in minor variations of soil com- 
position), or are not commonly recognized as important (as in the 
case of some light and temperature effects). This frequent difficulty 
of naturalization is one reason for the rather small percentage of 
alien species that actually enter into the composition of most wild 
floras in undisturbed tracts. Even of the numbers that may be able 
to propagate successfully and remain year after year in one spot, 
or sometimes increase their area by aggressive extension, few are 
known definitely to be permanent and able to persist in the absence 
of Man. Most lengthening of the floristic lists by aliens is probablv 
only temporary. 

This brings us to the other main reason, namely the need for Man's 
continued protection, behind the rarity of fully naturalized alien 
plants even relatively to the number of aspirants. A plant which 
has escaped from cultivation or a cultivated area on to some adjoining 
rubbish dump or otherwise disturbed tract, even if it manages to 
perpetuate itself there for years as many do, is far from attaining 
the status of full naturalization. In between mere escape and 
complete naturalization lie the various stages of success — including 
capabilities of spread, colonization, and possibly even the attainment 
of a dominant position. This last is often accomplished locally or 
sometimes extensively by plants which are effectively dispersed and 
rank in growth, as in the case of Fireweed (Epilobium angustifolhim 


agg.). But that is merely on territory where the native vegetation 
has been disturbed or destroyed : if this native vegetation is left 
alone to develop naturally, the aliens, however rank and aggressive 
they may once have been, will in most instances disappear within a 
very few years. Likewise have the plants which used to be trans- 
ported in ships' ballast largely disappeared, since the discontinuation 
of the dumping of such ballast, from the stations in which they 
formerly grew as aliens. The discontinuation of a road or railway- 
line is apt to have a similar effect, and even those aliens which have 
managed to spread from the immediate vicinity of the travelled track 
usually disappear when Man's influence is removed and the sur- 
rounding vegetation comes back into its own. 

All this does not mean that Man's influence in changing the dis- 
tribution of plants is other than enormous, but rather that it is in 
many instances merely temporary, as the plants involved have 
become only incompletely naturalized and certainly not lastingly 
acclimatized. It should also be remembered that, with Man's 
increasing mobility, for every disappearance of a plant from an area 
there is probably on the average at least one new introduction else- 
where, though in this connection particular plants tend to have their 
ups and downs. Even in those parts of the world, such as New 
Zealand and Hawaii, where the native plants have been largely ousted 
over considerable tracts by adventive aliens, this has happened only 
following disturbance of the native plant communities as well as 
importation of the aliens by Man or his domestic animals. There 
it is widely contended that removal of Man's influence would lead 
to a reversal of the situation through return of the natives whose 
stronger competition would ultimately oust the alien colonists. 
This matter of strength of competition is one of the most important 
in the life of organisms, and is often the key to the present-day 
distribution of plants as v/ell as to their potential ranges. Crop 
plants, sheltered and pampered as they are (and have usually long 
been accustomed to being), are notoriously weak in competition, 
and consequently rarely to be found in a truly naturalized state. 

The adaptations of plants to particular habitat conditions are 
varied and sometimes so precise as to remain unnoticed, yet sufficient 
to prevent the leading of an independent life. Often a mere slight 
change in environmental conditions will threaten the very existence of 
a species. For example, a Mexican species of Birthwort (Aristolochia) 
when transplanted to Java flowered abundantly but failed to bear 
fruit — not because of any lack of pollinators but because the climate 


there was too humid for its normal biol6gical development, the 
pistillate stage of each flower being over by the time it opened. As 
has been pointed out by WulfT in the work cited at the end of this 
chapter, if to such precise needs ' we add the unceasing struggle 
for existence and the competition with the indigenous vegetation, 
we should not be surprised at the relatively small number of those 
species introduced by man for cultivation or accompanying him in 
his migrations that became fully naturalized components of the local 
flora \ As an outstanding example it may be mentioned that, of 
the nearly one thousand alien species in the flora of Madagascar, 
it is claimed that only one, which is particularly easily dispersed, 
has gained a foothold in plant communities undisturbed by Man. 
Only such territory may be considered as ecologically fully occupied. 

For cultivation, land has in general to be cleared of native vegeta- 
tion and otherwise specially prepared. Such ousting of the native 
flora gives the adventives a chance to spread; so does less complete 
destruction of the vegetation, for example by domesticated animals. 
But once the cultivation or other disturbance is discontinued, there 
ensues a struggle between the alien and native plants which usuallv 
ends in victory for the latter and return to approximately the original 
condition. This is particularly noticeable and rapidly effected in 
the more favourable forested regions, whereas in some others, such 
as the drier grasslands, the breaking of the sod or other disturbance 
may so upset the ecological balance as to make its return extremelv 
slow or even problematical. Another interesting example of this 
appears to be afforded in some of the most favourable situations in 
southwestern Greenland. Here the clumps of Willows in many 
valleys are nowadays separated by grassy tracts (cf. Fig. 116) much 
as they presumably were at the time of the extinction of the Viking 
colonies and their pasturing Sheep several centuries ago : at least, 
the present writer has been unable, during hundreds of miles of 
wandering in those now uninhabited regions, to think of anv other 
explanation of a remarkable phenomenon. Nor have the tree 
Birches returned at all widely, either in those parts of Greenland 
or in Iceland, since the ' forests ' were decimated in the early centuries 
of the present millennium. 

The majority of really widespread weeds, such as those which 
qualify as semi-cosmopolites, tend to be collective species (such as 
the Common Dandelion, Taraxacum officinale s.l.) or to consist of 
numerous races (as in the Couch-grass, Agropyron repetis) adapted 
to diverse habitat conditions. The distinction between these two 


categories is largely a matter of degree and hence of opinion, the 
important feature from our point of view being that such ' poly- 
morphs ' are able to occupy a wide range of situations and hence, 
often, of regions. This is, however, chiefly where Man has dis- 
turbed the native vegetation. At the other extreme we have the 
highlv specialized ' monomorphs ', such as most cultivated strains, 
which for successful growth have to be given conditions within a 
very narrow range of amplitude. In such circumstances they may 
grow well enough year after year and seemingly indefinitely. But 
once Man's influence is removed and the coarser local indigenes 
are allowed to return to the area which is their normal heritage, 
such pampered cultivates will disappear with surprising rapidity and 
even aggressive weeds will usually fail within a very few years. Thus 
in the famous Broadbalk Wilderness of Rothamsted Experimental 
Station in southern England, according to Sir William Ogg (in litt. 
et inch), ' The Wheat plants on the strip . . . which was allowed 
to run wild survived for only four years ', being by then reduced to 
1 a few stunted plants . . . barely recognizable as cultivated Wheat \ 
Subsequently ' a dense growth of bushes and young trees ' developed, 
which soon ceased to include even the hardier wheat-field weeds. 

Some Herbaceous Crops and Their Areas 

Most of the important plant products on which Man's sustenance 
depends come from field or other herbaceous crops of short duration. 
The plants involved are usually special domesticated strains that 
have been so highly selected and long cultivated that they are unable 
to compete with natural vegetation — perhaps anywhere in the world 
— but, with Man's aid, they fortunately flourish sufficiently to enable 
him to maintain his position of supremacy. Human civilizations 
have largely developed in relation to the availability of suitable crops, 
in particular cereals, and there is altogether widespread inter- 
dependence between crops and Man. Communities living outside 
the cereal belts are often backward to this day. 

Whereas each of the various crops commonly had a single region 
of origin — as indicated, for example, in the works of DeCandolle 
and Vavilov cited at the end of this chapter — the main ones have 
usually become important through having their areas spread by Man 
into other regions. It has even been said that ' no world crop 
originated in the area of its modern commercial importance '. Not 
only are these regions, the present-day areas of particular crop 


plants, often virtually as extensive as climatic possibilities allow, but 
by special breeding and cultivation techniques Man is always 
endeavouring to extend the potential areas into new regions. This 
is notably true in the case of Wheat, the northern limit of which 
has been pushed farther and farther towards the Arctic in recent 
decades. It is chiefly with the currently attained areas of the most 
important herbaceous crops (as opposed to ' woody ' ones, treated 
afterwards) that the present section will be concerned. And whereas 
the major vegetational belts and hence natural plant distributions 
may to some extent have conditioned human migration in the past, 
it is largely the crop-growing potentialities of different regions that 
determine the density of human population today. This we shall 
see in the next chapter, though with the modern ease and efficiency 
of transport, particularly, this general conclusion tends to become 
less and less applicable in some areas of intense industrial or mining 

We will now indicate briefly on a world-wide basis the significance 
and chief areas of cultivation of some of the more important and 
familiar herbaceous crops ; ornamental ' flowers ' are apt to be even 
more widespread owing to the special care, often including develop- 
ment under greenhouse or other highly artificial conditions, that is 
lavished upon them. The chosen examples will be treated under 
eight main headings. 

(i) Grains — The principal grains occupy about one-half of the 
world's croplands and of them Rice (Oryza sativa) is probably the 
most generally important, being ' an indispensable food of over half 
the population of the world '. It replaces the other cereals as the 
staff of life in many tropical and subtropical countries, and in several 
of the most densely populated of these its cultivation is the chief 
agricultural industry. Although 95 per cent, of the Rice cultivation 
of the world is in the Orient, where the crop presumably had its 
origin far back in antiquity, Rice is now cultivated practically 
wherever in the tropics its usual needs of abundant moisture can 
be economically satisfied, its distribution affording a good example 
of that of a warm-climate crop {see Fig. 66). For the many types 
of lowland Rice, which have to be flooded during part of their 
development, are the ones grown almost exclusively ; relatively 
unimportant is ' upland ' or ' hill ' Rice, which can be cultivated 
in drier situations much like those favoured by other cereals. 

Wheats (Triticum vulgar e and other species) constitute the chief 
cereals of temperate regions and the ones most important to the 



white race nowadays. Their areas of origin are doubtful but were 
evidently diverse as regards the different forms, some of which 
appear to have been cultivated for at least 6,000 years. Nevertheless 
the predominance of Wheats is relatively modern, other cereals, or 
mixtures comprising maslin, having been previously more widelv 
used for bread. Wheats were probably developed by selection from 
weedy types and hybridization with other Grasses, but are still being 
improved today. They are grown under a wide variety of climatic 
conditions, including some tropical ones (in winter), and, like 
polymorphic weeds, are the more widespread because of their 
diversity. Nevertheless the general distribution of W'heats is mainly 
temperate, as indicated in Fig. 67. 

Less widespread and important are Barley (Hordeum vulgare s.l.), 
Oats (Avena sativa and other species), and Rye (Secale cereale), 
though the first of these is probably the oldest of our major cereals 
and possibly of all currently cultivated plants. It was widespread 
already in Neolithic times and was used for bread even before Wheat. 
Oats probably had a long history as a weed in fields of primitive 
Wheat before becoming a crop in its own right, while Rye, which 
was unknown before the Iron Age but is now the world's second 
most important bread crop, apparently originated as a grain-field 
weed in Asia Minor. It can be grown on poorer soils than other 
cereals. Owing to its greater winter hardiness and ability also to 
mature grain under less generally favourable conditions than the 
other cereals mentioned, Rye tends to be cultivated chiefly in 
mountainous regions and about the northern limit of the Wheat 
belt, being important chiefly in the cool-temperate parts of the 
northern hemisphere — cf. Fig. 68. However, Barley is able to 
mature in a shorter summer than the other cereals, and so is the 
only one which the writer has seen being grown successfully for 
grain north of the 70th parallel of latitude in Norw r ay. 

Maize (Zea mays) is the largest of the cereals. According to that 
foremost student of its history, Professor Paul C. Mangelsdorf of 
Harvard University (in lift. 1957), ' No wild ancestor is known with 
certainty, but fossil pollen believed to be that of wild Maize has 
been found at a depth of more than seventy meters below the present 
site of Mexico City. Other evidence points to cultivated forms of 
Maize originating on the eastern slopes of the Andes in South 
America. Maize cultivation goes far back in prehistoric times. 
Grains found in burial sites in Peru already represent several different 
varieties, indicating that the plant had been grown for many centuries 




before the period of the Inca civilization. Radiocarbon determina- 
tions of primitive cobs found in Bat Cave in New Mexico, indicate 
that this material was between 5,000 and 5,600 years old.' In spite 
of its numerous forms, Maize is a crop mainly of rather exacting 
requirements of considerable summer moisture in warm countries. 
It does not ripen far north ; indeed not many regions have the 
right combination of environmental conditions for the raising of 
Maize on a large scale. Most notable is the eastern half of the 
United States, which produces about half the world's crop, although 
there are other considerable centres of production in South America, 
southern Europe, and eastern Asia, as indicated in Fig. 69. Maize 
is used principally for feeding Hogs and other domesticated animals, 
but is also favoured as a vegetable. 

The various types of Millets, belonging to several different genera 
of Grasses, and the Sorghums, belonging to the genus Sorghum, should 
also be mentioned as widely grown for forage, grain, and many 
other purposes. The Sorghums have been cultivated in Asia and 
Africa since very early times, constituting a staple food for millions 
of native peoples. Latterly they have come to be grown in other 
tropical and warm-temperate regions, being particularly useful 
because of their ability to grow under dry conditions and actually 
withstand droughts. 

(2) ' Root ' Crops — Of these the Irish or White Potato {Solarium 
tuberosum) is the most widely important, having no rival as an 
efficient producer of food, especially in relatively moist and cool 
countries. Although the Potato's origin lay in the mountainous 
portions of South America, over 90 per cent, of world production 
is now in Europe, whose population has increased substantially as 
a result of its cultivation. In comparison with the other leading 
food-crops, the average annual world production during 1 934-38, x 
expressed in millions of metric tons, has been estimated as approxi- 
mately 233 for Potatoes, 167 for Wheat, 152 for Rice, 115 for Maize, 
65 for Oats, 52 for Barley, and 47 for Rye. Actually these figures, 
although interesting, are only fragmentary for some crops, and are 
moreover misleading in that Potatoes contain at least 78 per cent, 
of water, against an average of only about 13 per cent, for cereals. 
Consequently, the actual dry-weight food production of Potatoes in 
that period was only about 51 million tons, whereas for Wheat it 
was approximately 145 million tons, and for Rice and Maize also 

1 The last period for which the F.A.O. Yearbook (vol. IX, part 1, 1955) gives 
pertinent statistics for the U.S.S.R. 



considerably in excess of that of the Potato. Yet under suitable 
conditions of moist climate and rich but light soil, the Potato is 
able to supply considerably more human food per unit area than 
anv of the cereals, the world crop during 1934-38 being produced 
by about 22 million hectares as opposed to an estimated 168 million 
hectares under Wheat. Relatively to the following two so-called 
' root ' crops, the Potato is hardy, especially in some of its numerous 
forms, Fig. 70, A, indicating most of its cultivated range in the world. 
In addition the present writer has eaten quite large home-grown 
Potatoes in central Alaska and far down the Mackenzie Valley, and 
small ones in southern Greenland and in northernmost Scandinavia 
near 71 ° N. lat. He has even eaten tiny ones much farther north 
in Spitsbergen, grown on the pyre of a burned-out hut. In the 
White Potato the storage tuber is really an underground stem 
structure bearing buds (the so-called ' eyes '). 

The Sweet Potato (Ipomoea batatas), an ancient crop of tropical 
America, is now widely cultivated in the warm parts of the world, 
being in fact a standard article of food in practically all tropical 
and subtropical regions. It requires a sandy soil and a moist climate 
for successful growth. Another very important tropical food plant 
of this nature (though actually shrubby) is the Cassava (Manihot 
esculenta), which originated in South America in prehistoric times. 
It can be cultivated in hot, seasonally arid climates where cereals, 
etc., will not grow. Its many varieties now furnish the basic food 
for millions of people, particularly in Central and South America, 
and also supply the world with tapioca. 

Other important ' root ' crops which are widely used as vegetables 
include various types of Yams (Dioscorea spp.) in warm regions, and 
Turnips and Rutabagas (Swedes) which are also used for animal 
feed particularly in temperate regions. There are also Beets (Beta 
vulgaris) and Carrots (Daucus carota), which both succeed under a 
wide range of climatic and soil conditions. Beets were domesticated 
first as a leaf vegetable, then as root crops, and finally as a source 
of sugar (cf. upper part of Fig. 75) ; they are probably derived 
from one variable species that is native in the Mediterranean region. 
Carrots are likewise of ancient origin, various form, and now very 
widespread cultivation. 

(3) Other Vegetables — This somewhat vague category includes 
some structures (such as Tomatoes and the pods of Beans) which 
technically are fruits. Examples are the Broad Bean (Vicia faba), 
which is one of the world's commonest and most important beans 



Fig. 70, B. — Bed of young Kale, with, behind, tall Rhubarb, in Lichtenau Fiord, 

southwestern Greenland. The native family are standing on a raised path in 

front of the old mission house from which the garden slopes downwards. 

and was the only edible one known in Europe before the time of 
Columbus ; the Common or Garden or Kidney Bean (Phaseolus 
vulgaris), which has long been cultivated in the New World where 
it probably originated ; and the Soybean (Glycine max), which is 
of great antiquity in the Orient, where over one thousand varieties 
are grown. Soybean, particularly, has a very wide range of uses, 
the seed, containing about 20 per cent, of oil and 30-45 per cent, 
of protein, being the richest natural vegetable-food known. The 
climatic and soil requirements ideally are much like those for Maize, 
and the crop is becoming more and more extensively cultivated in 
temperate regions — including the United States, where it is grown 
chiefly as a source of oil and stock feed. 

Other important legumes are the Common Pea (Pisum sativum), 
which is now very extensively cultivated, and the Chick Pea (Cicer 
arietinum), which is an important food plant — particularly in India 
and other parts of Asia, in Africa, and in Central America. Both 
these plants appear to be natives of southern Europe or adjacent 
regions, where they have been grown from early days and are still 
extensively cultivated, and neither is known in the wild state. The 


Common Pea needs plentiful moisture but thrives in cool regions, 
whereas the Chick Pea is well adapted to dry conditions. Another 
widespread and important crop plant of this general affinity is the 
Lentil {Lens esculent a), which has been cultivated since Neolithic 
times and is thought to have originated in southwestern Asia. 

Mustards are extensively cultivated for their oil and use as greens 
in Asia, as are the related Cabbages and Kales and their allies 
(Brassica oleracea) in Europe and elsewhere for human and domestic 
animal consumption (though originally for their oily seeds). The 
Cabbages, etc., were evidently developed far back in antiquity from 
a variable Mediterranean species exhibiting numerous local races ; 
now they are grown practically around the world, occupying a wide 
variety of soils and climates ranging from the low-Arctic to the sub- 
tropics. Fig. 70, B, shows a fine bed of young Kale growing in 
southern Greenland, with, behind, tall Rhubarb. 

Finally we should mention a few ' fruit vegetables ' such as the 
Squashes and Cucumbers and their allies, many of which have been 
extensively cultivated from early times, and the widely important 
Tomato {Ly coper si cum esculentum), which springs from a group of 
small-berried weedy natives of Peru. Most of these types do best 
in warm and moist regions, to which some are practically confined. 

(4) Forage Plants — While several of the above-mentioned crops 
may be used in part for forage, there are some more specific forage 
plants to be mentioned primarily in this connection. Foremost 
among these are various Grasses, of which the Bluegrass or Meadow- 
grass (Poa pratensis s.l.) is an outstanding example. Cytotaxonomi- 
cally it is one of the most complex mixtures of polyploid hybrids-^ um- 
apomicts known, appearing in numerous forms whose origin is often 
obscure. Geographically it is widespread, particularly in the cooler 
regions of the world. Ecologically it is unexacting and aggressive, 
frequently forming a major constituent of pastures whether or not 
it has been eown. Another important and widely cultivated forage 
plant is Alfalfa or Lucerne {Medicago sativa), which was probably 
the earliest forage crop to be developed — apparently in southwestern 
Asia. Alfalfa prefers a deep, well-drained soil but is grown under 
a wide range of moisture as well as temperature conditions. It 
belongs to the Pea family as do also the Clovers and Vetches, which 
are themselves of considerable significance in pasturage and hay. 

(5) Fibre and Oil Plants — In this extensive category the Cottons 
(various species of Gossypium), Flax (Linum usitatissimum), Hemp 
{Cannabis sativa), Jute (species of Corchorus), and Peanut or Ground- 


nut {Arachis hypogaea) are outstanding in importance and widely 
cultivated. The first three are used chiefly for fibre and oil, Jute 
for fibre, and Peanuts for oil and food. All are plants primarily 
of warm regions, Flax alone being successful far to the north (cf. 
Fig. 71). Cotton as a whole is often claimed as the world's greatest 
industrial crop and chief source of fibre ; its multiple origin is 
shrouded in the mists of time. The world distribution of Cotton 
production is indicated in Fig. 72. The others, too, are of ancient 
and often uncertain origin — mostly in the Old World, but the Peanut 
very likely in South America, though it is now extremely widespread 
(see Fig. 73). 

Other important vegetable fibres are Ramie or China-grass 
(Boehmeria nivea, widely cultivated in Asia), Sunn-hemp (the Asiatic 
Crotalaria juncea), the chiefly Philippine Abaca or Manila-hemp 
(Musa spp.), Sisal and other Agave types, cultivated in Africa and 
North and Central America, and filling fibres such as Kapok (the 
floss from the seeds of the now widespread tropical tree Ceiba 
pentandra). Of oils there are the essential or volatile types used 
particularly in perfumery, and the fatty or fixed types which include 
the drying (e.g. Tung, from species of Aleurites, native to China), 
semi-drying (e.g. the Asian Sesame), and non-drying (e.g. Olive) 
categories. Olives are cultivated chiefly in the Mediterranean lands 
but to some extent also in the United States, South Africa, and 
Australia. In addition there are the vegetable fats such as palm oil 
(obtained from the African Oil Palm, Elaeis guineensis) and coconut 
oil (obtained from the Coconut — see pp. 242, 266). Of the sources 
mentioned above, cottonseed oil, obtained from Cotton, is the most 
important semi-drying oil, linseed oil, obtained from Flax, is an 
important drying oil, and hempseed oil, obtained from Hemp, is 
another, while peanut oil is a non-drying oil. 

(6) Fruits — Whereas, technically, many of the above-mentioned 
products are fruits or derived from fruits, the term is used here in 
the popular sense. Most of our main fruits are borne by trees or 
shrubs and will be dealt with in the next major section. The 
Pineapple (Ananas comosus) and Melon (Cucumis melo), and, for all its 
appearance, the Banana (Musa par adisiaca s.l.), are, however, strictly 
herbaceous, as are Strawberries and some other favourites. The 
cultivated Strawberry (Fragaria grandiflora) is, according to Pro- 
fessor Edgar Anderson, ' the one crop of world importance to have 
originated in modern times ' — actually in the eighteenth century as 
a true-breeding polyploid hybrid from artificial crosses between wild 



2 3 8 


North and South American types. Other Strawberries have long 
been grown elsewhere, the fruit as a whole being a favourite in all 
temperate countries. In contrast, Pineapples, Bananas, and Melons 
are mainly tropical types of ancient origin. The Malay Peninsula 
appears to have been the chief centre of origin of cultivated Bananas, 
whereas South America gave us the Pineapple, and Africa or southern 
Asia various Melons ; but all three types are now grown practically 
around the globe. 

(7) Other Crops — Especially important among these are Tobacco 
(Nicotiana tabacam) and Sugar Cane (Saccharum officinarum), while 
the Hop (Hamulus lupulus) and various Buckwheats (Fagopyrum 
spp.) are of no mean significance in some areas. Tobacco is 
apparently a true-breeding polyploid hybrid between two weedy 
inhabitants of South America, where it probably arose in cultivation 
in early pre-Columbian times. Now it is grown extensively in 
various of the sufficiently summer-warm regions around the globe, 
as indicated in Fig. 74, and is an important commodity throughout 
the inhabited world. Sugar Cane, a vigorous-growing perennial 
Grass, is the chief source of sugar at present, although at times in 
the past it has been rivalled by Sugar Beets. Sugar Cane probably 
originated in southeastern Asia and comprises an assemblage of 
forms that are unknown in the wild state but are now cultivated in 
practically all moist tropical and subtropical regions, the main areas 
of production of sugar from it and from Sugar Beets being indicated 
in Fig. 75. Cane sugar probably constitutes the greatest export 
crop of the tropics. 

(8) Raw Materials for Industry — A large proportion of these are 
afforded by plants in limitless supply. This category largely cuts 
across the others, which in most instances contribute familiar 
examples to it, and so we need scarcely add details. Suffice it to 
say that industrially important raw materials include not only the 
examples already mentioned, such as various grains, roots, fibres, 
oils, carbohydrates and their derivatives, but also a wide range of 
forest products including rubber and pulp. The field-crops involved 
are grown in the main crop-producing parts of the world and the 
forest products are obtained mostly in major forested regions. 
Further information about the sources of these all-important raw 
materials supplied by plants will be found in the books cited at the 
ends of this chapter and the succeeding one which stresses further 
their vital economic significance. 



242 introduction to plant geography [chap. 

Forestry and Other Woody ' Crops ' 

The practices of forestry, being largely directed towards more 
effective utilization of the forested regions of the world, are to a 
large degree concerned with cropping. Especially when artificial 
planting of trees or shrubs is involved — often with marked effect 
on their natural ranges, and sometimes with the maintenance of 
special strains — do the results demand some consideration here. 
Woody plants as a whole greatly extend the category of fruits (No. 6) 
introduced above, add one of beverages, and above all contribute 
their own vast one of timbers and cognate products with which we 
can deal only in brief outline. 

Trees and other woody plants that are extensively cultivated for 
edible fruits of importance to mankind include the Coconut (Cocos 
nucifera), Breadfruit (Artocarpus altilis), Olive (Oka europaea), Date 
Palm (Phoenix dactylifera), Fig (Ficus carica), Citrus fruits (Citrus 
species), Grape (Vitis vinifera), Currants and Gooseberries (Ribes 
species), Mango (Mangifera indica), Papaya (Carica papaya), Plum 
(Prunus domestica and other species), Peach (Prunus persica), Apple 
(Pyrus mains), and Pear (Pyrus communis). The first three are 
scarcely fruits in the lay sense. 

The Coconut is sometimes claimed to be the most important or 
at all events thoroughly exploited of cultivated plants, being used 
also as a source of timber, thatch, fibre, and many other things — 
especially by primitive peoples, who may be almost wholly dependent 
upon it and use all parts. A native of southeastern Asia, where 
the wild trees are still cropped, it has been carried to practically all 
tropical and subtropical shores, being very extensively planted. 
Another important Palm is the Date, which is widely cultivated in 
the tropics and subtropics where it can be grown with less water 
than any other crop. It is one of the oldest of crops and is supposed 
to have originated in southwestern Asia, though it is unknown in 
the wild state. The Breadfruit, a native of Malaya that is now r 
widespread in the tropics, having been cultivated from earlv times, 
is another very important food fruit (in the botanical sense). A 
Man is said to be able to live throughout the year on the products of 
a single tree. Most of the remaining fruits mentioned are attractive 
and familiar cultivates that have long been widespread in the climatic 
belts which they favour, ranging from the cool-temperate Currants 
and Apples to the tropical Mango and Papaya. Here we might add 
the Mangosteen (Garcinia mangostana), which is regarded by some 
as the world's most delectable fruit, 


Important beverages obtained from woody plants include Cacao 
(Theobroma cacao), the source of cocoa and chocolate, Coffee (various 
species of Coffea), and Tea {Camellia sinensis). The Cacao tree is 
a native of tropical America ; the others originated in the warm 
parts of the Old World. All are now extensively cultivated in the 
tropics, and, in the case of Tea, in other warm regions. Fig. 76 
indicates the main centres of production of Cocoa beans and 
exemplifies a tropical crop of restricted origin that is now widespread. 
Fig. 77 gives similar indications for Coffee, over half of which comes 
from Brazil, whose economy is still bound up with this single crop 
to an economically unhealthy degree. 

Passing over further categories such as nut-bearing, rubber, and 
drug plants, whose important products are often obtained from wild 
sources, we come to the last great one of timbers and cognate forest 
products. For details of these, reference may be made to such 
works as that of Zon & Sparhawk or, for the New World, of Record 
& Hess, both of which are cited at the end of this chapter. 

Besides the timbers employed almost all over the world for con- 
struction, fuel, and other purposes, important forest products include 
tanning and dyeing materials and a great assortment of useful gums, 
resins, oils, preservatives, cork, and latex products — to name only 
a few. Many of these are taken with fair regularity as a kind of 
crop, sometimes from planted trees. And whereas in the tropics 
the vast array of generally mixed timber trees are usually of rather 
restricted distribution, the relatively few types occurring in temperate 
and boreal regions are often widely transported and cultivated. 
Good examples are found among the Conifers that are successfully 
planted in Europe, which is deficient in native trees for reasons that 
were discussed in Chapter VI. Such Conifers have often been 
transported from North America (as in the case of the Douglas 
Fir, Pseudotsuga taxifolia) or Asia (whence come especially numerous 
ornamental types, though admittedly these scarcely constitute crops), 
but do quite well at least as long as Man's influence prevails. Apart 
from such artificial introduction, there are very few large woody 
species common to both sides of the Atlantic — in contrast to the 
situation with numerous herbaceous species especially in the boreal 
and arctic regions. The tree genera, however, are commonly the 
same in Europe and the temperate parts of North America and eastern 
Asia, though some have disappeared from Europe in recent geological 

Forests occupy about one-quarter of the total land-area of the 




world, as indicated in Fig. 65, which also suggests that a roughly 
similar proportion is occupied by each of the other three main types 
of landscape, namely, grassland (with savanna), desert or semi-desert, 
and tundra (with fell-field, etc.). It has been estimated that South 
America has about 44 per cent, of its land area forested, Europe 
about 31 per cent., North America about 27 per cent., Asia about 
22 per cent., Africa about 11 per cent., Australia about 6 per cent, 
(although New Guinea has 80 per cent.), while Antarctica has no 
forests at all. In many countries the forests were formerly much 
more widespread than they are today, the reduction being due 
primarily to interference by Man, but in some of the more civilized 
lands extensive reforestation is now being undertaken. This 
planting is often of exotics introduced from distant regions of com- 
parable climate, the plantations representing a kind of crop whose 
range is thereby greatly extended. 

The characteristics of the main types of forest will be described 
below in the appropriate chapters on vegetation. Here it will suffice 
to mention a few of the more important timber trees which in most 
instances are widely planted and tended (and to that extent, as well 
as in their regular use by Man, qualify as crops). In North America 
nowadays the Yellow Pines, Douglas Fir, Hemlocks, White Pine, 
Cypress, and Spruces tend to be the most important softwoods, 
with Oaks, Red Gum, Maples, Birches, and Poplars leading the 
hardwoods, of which the area occupied and the annual ' cut ' are 
much smaller than in the case of softwoods. 

In Europe, 74 per cent, of the forests are classed as coniferous, 
and such forests, as in America and Asia, are particularly char- 
acteristic of the northern portions. The principal European Conifers, 
which are frequently grown in special plantations, are the Scots 
Pine (Pinus sylvestris), Norway Spruce (Picea abies), and Larch 
(Larix decidud), though the American Douglas Fir and certain other 
Pines are extensively planted. The most important European hard- 
woods tend to be certain Oaks, but Beech (Fagus syfratica), Ash 
(Fraxinus excelsior), and some Birches and Elms are also prominent. 
The genera are thus much the same as in North America although 
the native species are different, and this situation continues over 
much of northern Asia. Here, in the west, European species are 
found, but these tend to give way to Asiatic species of the same 
genera farther east. Conifers comprise an estimated 42 per cent, 
of the forest area of Asia, and temperate hardwoods 27 per cent., 
the remainder being made up of tropical hardwoods which in many 


countries south of the Himalayas comprise nearly 100 per cent, of 
the trees. The species of tropical hardwoods in an area are often 
very numerous, India, for example, being estimated to have fully 
2,000. The dominance tends to be intricately mixed, although Teak 
(Tectona grandis) and various members of the Dipterocarp and Pea 
families are often prominent. Important commercially, if not 
ahvavs ecologically, are Ebony (various plants including Diospyros 
ebenum), Satinwood (Chloroxylon swietenia), and Burmese Rosewood 
(Pterocarpus indicus). 

South America, as we have seen, bears a greater proportion of 
forested area than remains on any other continent. Nearly 90 per 
cent, of its forest is tropical hardwood — mainly dense rain forest 
which characterizes the great river basins and tends to be very 
luxuriant and intricately mixed (there are said to be over 2,500 
different tree species in the Amazonian forests alone). In some 
drier areas an open deciduous type of tropical forest occurs, and on 
the high mountains are mixed forests of Conifers and temperate 
hardwoods. Important woods of tropical America include Balsa 
(Ochroma lagopus s.L, the lightest of commercial timbers), Spanish- 
cedar (Cedrela odorata s.L, forms of which are native in some areas 
but introduced in others), Greenheart (Ocotea rodioei), Lignum- 
vitae (species of Guaiacum), Locust (Hymenaea courbaril), and 
Mahogany (chiefly Swietenia macrophylla and S. mahagoni, of which 
the latter has been widely introduced). 

In Africa, contrary to popular conception, forests cover only about 
11 per cent, of the land area. Tropical hardwoods predominate, 
comprising some 97 per cent, of the forests. Here again there are 
two main types, of which the dense and much-mixed rain forest is 
the more extensive but is replaced by an open, park-like type where 
the rainfall amounts to only 30-40 inches per annum. Of the woods 
that have so far been exploited, an outstanding example is the 
African Mahogany (Khaya senegalensis), which is widely exported. 

Although in Australia forests cover only a very small proportion 
of the land area, in New Zealand the percentage is about 26 and in 
Oceania 71. In Australia tropical hardwoods predominate — par- 
ticularly species of Eucalyptus and Acacia — and in Oceania they make 
up the entire forest. On the other hand in New Zealand 68 per 
cent, of the forests are coniferous and the remainder temperate 
hardwoods, though the genera tend to be different from those pre- 
dominating on other continents, 

248 introduction to plant geography [chap. 

Significance and Distribution of Weeds and Plant Diseases 

Although there are many tens of thousands of species of higher 
plants in the world, only a few dozens of these are really troublesome 
weeds which are able to reproduce and thrive in the presence of 
cultivation and other interfering human activities. More numerous 
by far are the weedy plants and ' escapes ', both herbaceous and 
woody, which for the present purpose may be considered with the 
more noxious weeds, and which link the latter with the categories 
of cultivated plants. Yet the annual loss due to weeds is enormous, 
often amounting to millions of dollars in a relatively small area. 

Weeds are injurious to agriculture, e.g. by robbing crops of 
needed water and nutrients, by crowding them out through root- 
competition or overgrowing, by choking and pulling them down in 
the case of (sometimes parasitic) climbers, by having seeds or fruits 
so similar to those of the crop that they are difficult to separate and 
so adulterate it and reduce its value, by harbouring undesirable 
insects or plant diseases, by being poisonous or injurious to stock, 
by tainting milk, and so on. The nuisances caused by Bindweeds 
and Couch-grass are all too familiar to almost every gardener as 
well as farmer in the temperate belt ; the Prickly-pears (species of 
Opuntia) which were introduced into South Africa and Australia as 
a stock feed now usurp the ground ; Barberries and Currants 
harbour (as alternative ' hosts ') the devastating Wheat Stem-rust 
and White Pine Blister-rust, respectively (see pp. 251-2). 

Regardless of the common ' effects of cultivation ' listed near the 
beginning of this chapter, annual weeds often produce numerous 
seeds whose germination may be distributed over many years — for 
example, after being buried for decades. 1 Moreover, the seeds or 
fruits of many weeds, such as Thistles and Dandelions, are provided 
with efficient means of dispersal. Otherwise their wide distribution 
seems to be largely due to Man's transportation activities — such as, 
for example, the shipment of commercial seeds and grain. These 
are often of much the same size and shape as the disseminules of 

1 The longevity of seeds and fruits comprises an interesting study for which 
more and more authentic data are needed. Discounting claims of longer periods 
of burying in marshes etc. which are not fully authenticated, and stories of 
1 mummy Wheats ' which are clearly bogus, the record seems to be held by a 
' seed ' reputedly of Nelumbo nucifera (syn. Nelumbium speciosum, the Sacred or 
Indian Lotus) which was germinated in the British Museum (Natural History), 
South Kensington, London, during the bombing of 1940, apparently about 250 
years after it had been collected. 


the weeds which adulterate them, and which are apt all too easily 
to be shipped and sown with them. It is apparently largely in this 
manner — though further means were listed in Chapter IV — that, 
among others, the following pernicious European weeds have become 
dispersed to temperate America, South Africa, New Zealand, and 
temperate Australia : Couch-grass (Agropyron repens), Crab-grasses 
(species of Digitaria), Russian-thistle (Salsola kali var. tenuifolia) y 
Bindweeds (species of Convolvulus), Sheep Sorrel (Rumex acetosella 
agg.), Dodders (species of Cuscuta), Plantains (species of Plantago), 
Wild Carrot (Daucus carota), Prickly Lettuce (Lactuca scariola) y and 
Sow-thistles (species of Sonchus). Europe is not, however, by any 
means the only source of weeds : although it has contributed some 
500 to North America alone — including all present-day Americn 
representatives of Lamium, Melilotus, Medicago, Maha> and some 
other familiar genera — many weeds have been dispersed in the 
opposite direction. Some have even gone much farther afield than 
across the North Atlantic — e.g. Canadian Fleabane (Erigeron 
canadensis) and Gallant Soldier (Galinsoga parviflora y a native of 
South America). 

These and others among the most widespread of weeds are the 
' semi-cosmopolites ' mentioned in the last chapter, and it is notice- 
able that some of them lack special means of dispersal — commerce 
having gradually distributed them to all the major regions where 
they can thrive. This is notably the case with such other types as 
Shepherd's-purse, Common Chickweed, Annual Meadow-grass, and 
Lamb's-quarters (Chenopodium album s.l.), which are among the 
most widespread of all flowering plants because, primarily, of Man's 
unwitting transport and, secondarily, of their variability which 
enables them to grow in a wide range of different habitats. How- 
ever, as we have already seen, this is usually manifest only so long 
as Man continues to interfere by clearing areas or at least keeping 
the native vegetation at bay ; competition tends to be too much 
for weeds which lack Man's help. This help often extends to 
breaking up underground parts from which vigorous regeneration 
can take place — for example in the case of the rhizomes of Couch- 
grass and Bracken, and the roots of Dandelions. Such weeds are 
particularly difficult to eradicate. 

Although the semi-cosmopolitan weeds tend to have such a wide 
tolerance to environmental conditions that they are capable of 
invading almost any agricultural or otherwise disturbed area, at least 
within their normal climatic limits, other weeds are far more exacting 


and restricted. Consequently some regions have a characteristic 
weed-flora of their own, to which many of the component species 
are largely limited. Thus whereas many weeds have nowadays the 
type of distribution exhibited by some crops, that is, practicallv 
world-wide within climatically limited bounds, others are restricted 
by special conditions to particular areas for which they are suitable. 
In the former case peripheral portions of the distribution are often 
very temporary besides being artificial, Man having extended the 
area far beyond its natural bounds. Moreover, as we might expect 
from examination of the ranges of wild plants, weeds in areas outside 
those of their climatic optimum tend to be restricted to particular 
habitats where such factors as soil type or microclimate compensate 
for unfavourable climatic conditions. Here even vigorous weeds 
may be only sporadic in appearance, easy to control, and liable to 
disappear quickly in the absence of human interference. Neverthe- 
less, many of the weeds which follow particular crops are almost as 
widespread as the crops they accompany. 

Modern plant pathology, the scientific study of plant diseases, is 
a large and important subject in its own right — as may be gathered 
for example by perusal of such works as those of Butler & Jones, 
Walker, and Boyce cited at the end of this chapter. Only a few 
selected distributional and allied items can be touched upon here, 
for almost any abnormal state of plants, such as may depress the 
yield of a crop, is liable to be considered as a disease. 

Plant diseases may conveniently be classified into three primary 
groups : (1) non-parasitic, incited primarily by such physical or 
chemical factors as low or high temperatures, unfavourable oxygen 
or soil-moisture relations, atmospheric impurities, lightning, and 
mineral and other excesses or deficiencies ; (2) parasitic, incited by 
Bacteria, various groups of Fungi and their allies, Angiosperms, 
and animals such as insects and Nematode worms ; and (3) virus 
diseases. Examples of most types can be lethal — sometimes to all 
the plants belonging to a particular species throughout a tract of 
country. They are therefore of great importance to plant distribu- 
tion ; for they are apt in some cases to attack any part of the area 
of a particular species or even wider taxon, and consequently to be 
of the utmost significance to mankind whose crops they so frequently 
affect adversely or ruin or even basically destroy. 

Whereas very numerous and often serious diseases are known of 
wild plants throughout the world, it is chiefly among cultivated ones 
that the worst ravages are caused or at all events noted. Here 


epidemics are of frequent occurrence and sometimes vast propor- 
tions. For by growing crops in close plantations, Man offers to 
parasites a great opportunity for rapid growth and reproduction, 
while non-parasitic diseases which affect a particular crop plant may 
be strikingly evident owing to the absence of other species which 
might mask the effect. 

Although Bacteria and viruses cause many serious plant diseases, 
bv far the most important group in this connection are the Fungi. 
They may range from very local to very widespread in distribution, 
sometimes covering virtually the whole area occupied by the plant 
attacked (commonly called the ' host '). Indeed it seems quite likely 
that particular diseases have been responsible for the complete 
extermination of some plants in the past, even as they can nowadays 
cause the disappearance of particular plants from considerable areas. 
This can obviously be of vast significance in plant geography. As 
an example we may cite the notorious Chestnut Blight which in recent 
decades has almost exterminated Sweet Chestnut trees from the 
United States, where formerly they were of major importance both 
ecologically and economically. Another example of an important 
plant disease is Late-blight of Potato, which led to the great Irish 
famine of the eighteen-forties that resulted in the deaths of hundreds 
of thousands of people and started the wholesale Irish peasant 
migration to the United States. Yet another is Wheat Stem-rust 
that has caused an estimated loss in western Canada alone of as 
much as $200,000,000 in a single year. In the tropics, another 
Rust Fungus caused the disappearance from Ceylon of the Coffee 
industry which had long been the mainstay of its prosperity, and 
further instances could be cited of such epidemic plant diseases, 
often introduced from afar, profoundly influencing the economic 
development of a country, or causing enormous loss or acute distress 
over considerable areas. 

The examples mentioned are all of airborne pathogens, dispersed, 
in part at least, as minute spores which are blown by the wind often 
for considerable distances. Accordingly such diseases as the cereal 
Rusts and similarly airborne Smuts are present in all cereal-producing 
countries. The chief means of combating them is by the breeding 
and cultivation of resistant or immune strains, or, in the case of the 
more intensive crops, by poisonous sprays, etc., which are lethal 
to the infecting organism. In other instances suitable treatment of 
contaminated soil or seed, or eradication of infected plants, will 
destroy the parasite. Or again, the imposition of strict quarantine 


barriers can be effective, while with some virus and other diseases 
which are spread by insects, the killing of these vectors, or growth 
of the crop in areas where they do not occur, should suffice for 
successful control. Some plant pathogens have an alternative host 
in which part of the life-cycle is spent. In such cases eradication 
of this alternative host is effective, an example being afforded by 
the White Pine Blister-rust, which threatens the life of all five- 
needled Pines in North America, and whose alternative host is the 
genus Ribes (Currants and Gooseberries). It should be noted, 
however, that such systematic eradication may not merely affect the 
distribution of the species concerned but also the local ecological 
balance, which may be seriously upset. Similarly, DDT spraying 
against harmful insects may at the same time kill off all the bees 
which are necessary for pollination! 

Further Consideration 

E. V. Wulff. An Introduction to Historical Plant Geography (Chronica 
Botanica, Waltham, Mass., pp. xv -f- 223, 1943) ; for further details 
about many of the topics discussed in the early part of this chapter. 

Origins of Crops : 

Alphonse DeCandolle. Origin of Cultivated Plants (Kegan Paul & 
Trench, London, pp. ix + 468, 1884) ; still useful. 

N. I. Vavilov. The Origin, Variation, Immunity and Breeding of 
Cultivated Plants, translated by K. S. Chester (Chronica Botanica, 
Waltham, Mass., vol. 13, Nos. 1-6, pp. xviii -f- 364, 1951) ; see also 

Edgar Anderson. Plants, Man and Life (Little, Brown, Boston, Mass., 
pp. [vii] + 245, 1952) ; stimulating. 

E. D. Merrill. The Botany of Cook's Voyages (Chronica Botanica, 
Waltham, Mass., vol. 14, Nos. 5-6, pp. i-iv + 161-384, 1954) ; for 
additional information, with a master's pungent criticisms of some 
previous contentions. 

For Details about Herbaceous Crops : 

E. E. Stanford. Economic Plants (Appleton- Century- Crofts, New York, 

pp. xxiii + 571, 1934). 
K. H. W. Klages. Ecological Crop Geography (Macmillan, New York, 

pp. xviii -f- 615, 1942). 
A. F. Hill. Economic Botany, second edition (McGraw-Hill, New York 

etc., pp. xii I 560, 1952). 


L. H. Bailey et al. Manual of Cultivated Plants, revised edition (Mac- 
millan, New York, pp. 1-1116, 1949). 

For Details about Forest Products : 

R. W. Schery. Plants for Man (Prentice-Hall, New York, pp. viii -f 564, 

R. Zon & W. N. Sparhawk. Forest Resources of the World (McGraw- 
Hill, New York & London, 2 vols, pp. xiv 4- 1-493 and vii -f- 495- 

997> x 9 2 3)- 
S. J. Record &: R. W. Hess. Timbers of the Nezv World (Yale University 

Press, New Haven, Conn., pp. xv -f 640, 1943). 

S. Haden-Guest, J. K. Wright, & E. M. Teclaff (ed.). A World 

Geography of Forest Resources (Ronald Press, New York, pp. 

xviii - 736, 1956). 

Weeds and their Control : 

L. J. King. World Encyclopaedia of Weeds and their Control (Leonard 

Hill, London, in Press). 
W. C. Muenscher. Weeds, second edition (Macmillan, New York, pp. 

xvi — 560, 1955) ; mainly northern United States and southern 

W. W. Robbins, A. S. Crafts, & R. N. Raynor. Weed Control, second 

edition (McGraw-Hill, New York etc., pp. xi + 5°3> 1 9S 2 '). 

Plant Diseases : 
Sir E. J. Butler & S. G. Jones. Plant Pathology (Macmillan, London, 

pp. xii - 979, 1949). 
J. C. Walker. Plant Pathology, second edition (McGraw-Hill, New 

York etc., pp. xi + 707, I 957)- 
J. S. Boyce. Forest Pathology, second edition (McGraw-Hill, New York 

etc., pp. xi + 550, 1948). 

In conclusion it may be interesting to speculate as to the main centres 
of origin of our crop plants. It has long been thought that the cultivation 
of plants by Man began independently in each of the three main centres 
of ancient civilization, viz., the eastern Mediterranean, the Oriental of 
southeastern Asia, and the American of southwestern North America 
and at least the northwestern portions of South America. Nowadays 
there is a tendency to extend or multiply these to include other areas of 
supposed early cultivation. Vavilov (op. cit.) visualized eight major 
centres, which he found to be those of greatest diversity of cultivated 
plants, as follows : (i) Chinese, (ii) Indian (with a suggested separate 
Indo-Malayan area), (iii) central Asiatic, (iv) near-Eastern, (v) Mediter- 
ranean, (vi) Abyssinian, (vii) south Mexican and Central American, and 


(viii) South American (Peruvian-Ecuadorian-Bolivian and two minor 
areas). Much further extensive as well as intensive investigation should, 
however, be carried out before (if ever) broad generalizations may be 
indulged in where so many intangibles are involved ; definite information 
is still largely lacking. This is why such a fascinating and potentially 
important subject does not receive more consideration in this book. 
But it does seem that cultivated plants mostly originated in warm regions, 
if often in their upland areas. 

As this book is in the press, there comes news of the discovery in Jericho 
of a Neolithic culture considerably older than any formerly known. All 
possibility of observing signs of the ancient cultivation which must have 
existed around the site has been destroyed by modern agriculture. But 
the large extent of the area of sedentary occupation constitutes reasonably 
certain proof that agriculture of some sort must have been prosecuted 
at that time, which on the evidence of carbon- 14 dating was around 
7000 B.C. (K. M. Kenyon voce). Among other plants, Linum appears to 
have been cultivated in the plains of Iraq as early as around 5000 B.C. 
(H. Helbaek voce). 

Chapter IX 

Early in this work we indicated that the green plant is the only 
satisfactory mechanism for transforming the energy of the sun into 
organic compounds on which, as an animal, Man is dependent for 
food and other requisites of life. In the last chapter we mentioned 
some instances of such dependence, chiefly in connection with the 
distribution of leading crops. It is the object of the present chapter 
to give a systematic account, with chosen examples, of the multi- 
farious and often vital ways in which plants and plant products are 
important to mankind. For Man is unable to synthesize, at all 
events economicallv and in useful bulk, most of the materials which 
he needs in such great quantities — often for his very existence. 
Even though he can, for example, convert starch into alcohol and 
the latter in turn into all manner of useful products, he needs the 
Wheat or some other plant to make the starch for him. In such 
food materials is locked radiant energy from sunlight, which can 
then be liberated by the process of respiration that goes on in all 
living bodies and is rapid in warm-blooded animals. This process 
usually requires oxygen in large quantities and consequently is again 
dependent upon green plants as, during photosynthesis, they liberate 
this vital gas and return it to the air, so purifying and maintaining 
the atmosphere. 

Not only, as we shall see, do plants afford for mankind, either 
directly or indirectly, his food and many other requisites of life, 
but they largely condition his environment. Thus, for example, 
forests are very different to live in from grassy plains, and deserts 
and areas of arctic tundra are again widely different. Many present- 
day grasslands and treeless cultivated areas are, however, due to 
Man's clearance of forests, and although he shows a natural tendency 
to avoid desert and tundra areas, the correlation of forest or grass- 
land {see Fig. 65) with dense human population (Fig. 78) is not 
always close. Rather has Man wandered and settled where he most 
conveniently could, having in mind his need for subsistence, which 
meant in large measure the finding or growing of plants — or of 


/S '•- 




animals which are dependent upon them. Thus the migrations of 
Man have been dependent in considerable degree on plant distribu- 
tion, even as his present population-density is conditioned by the 
crop-growing potentialities of different areas. 

From the point of view of what they are used (or, occasionally, 
to be avoided) for, plants and plant products may be classified into 
seventeen main (or often multiple) categories whose consideration 
will occupy the remainder of this chapter. As each is apt to be a 
large subject, the accounts will be brief or in mere outline ; it should 
also be noted that the categories are neither hard and fast nor 
mutually exclusive. Indeed a good deal of repetition is inevitable. 
Many of the more important plants are dealt with elsewhere in this 
work, though usually in other connections, and some are mentioned 
in more than one category — though without cross-referencing, as this 
can be done through the index. Nor, in this primarily ( economic ' 
chapter, will technical botanical usages and names be maintained 
in the manner in which they usually are elsewhere in this book, 
and practically have to be for the use of scientists. 


Though whole volumes can be — and have been — written about 
food plants and Man's dependence upon them, this theme is too 
obvious to require detailed treatment here. It has latterly been 
extended to include vitamins, of which plants are the main primary 
source. Suffice it to say that practically every item of food of all 
animals comes from plants. For even if one animal eats another, 
and it in turn consumes yet another, when we follow back the food- 
chain we come, sooner or later, to a point of dependence upon green 
plants, as these alone are able economically to build up complex 
food substances from simple inorganic materials. This is true not 
only on land but also in fresh and salt waters, where the ' producer ' 
plants are often microscopic, larger and larger ' consumer ' animals 
succeeding one another to constitute the later stages in the food- 
chain. Any exceptions are insufficient in scale to have serious effect. 

The most important plants used directly for food by Man (as 
opposed to those used indirectly through pasturage of his domestic 
animals) are those affording abundant carbohydrates and other 
energy-producing materials. Outstanding are the cereal and other 
1 grains ' (such as Wheat, Rice, Maize, Barley, Rye, Oats, Sorghums, 
Millets, Quinoa, and Buckwheat) and ' roots ' (such as Potatoes, 


Sweet Potatoes, Yams, and Cassava), on one or other of which 
practically the whole human population of the world is primarily 
dependent. Useful additional ' vegetables ' include Beets and 
Chard, Salsify, Carrots, Parsnips, Radishes, Swedes, Turnips, 
Jerusalem Artichokes, Taros, Dasheens, Onions and their allies, 
Artichokes, Asparagus, Cabbages and Kales and their allies, Celery, 
Chicory, Endive, Lettuces, Rhubarb, Spinach, Dandelions, Water- 
cress, Avocado, Breadfruit, Jack-fruit, Cucumber, Pumpkins and 
Squashes and their allies, Yautias, Chayote, Egg-plant, Okra, Tomato, 
etc. There is no need to point out that the staples among the above 
constitute the mainstay of human life on earth, different ones in 
different regions affording the chief (and often almost sole) food of 
teeming millions. 

Other important groups of foods are legumes (the fruits of members 
of the Pea family, Leguminosae) and nuts (the term being used in 
the layman's sense). The former category includes Peas of various 
kinds, Chick Peas, Pigeon Peas, Cowpeas, Beans of various kinds 
(produced by several different genera), Soybeans, Peanuts, Lentils, 
Lablab, Algaroba and other Mesquites, Carob and other Locusts. 
The nuts include some with a high carbohydrate content, such as 
Chestnuts and Acorns, others with a high protein content, such as 
Almonds, Beechnuts, and Pistachio-nuts, and many more with a 
high fat content, such as Coconuts, Brazil-nuts, Cashew-nuts, 
Hazel-nuts, Macademia-nuts, Hickory-nuts and Pecans, Walnuts 
and their allies, Pili-nuts, and Pine-nuts. 

Fruits used regularly by Man for food, and commonly cultivated, 
are very numerous as well as various in their botanical significance. 
Those produced mainly in temperate regions include such ' stone ' 
fruits as Plums and Prunes, Cherries, Peaches and Apricots, such 
1 pome ' fruits as Apples, Pears, and Quinces, such ' gourd ' fruits 
as Melons and Watermelons, and such ' berries ' as the various kinds 
of Grapes, Blackberries and Raspberries, Blueberries and Huckle- 
berries, Cranberries, Currants and Gooseberries, Strawberries, and 
Mulberries. The fruits produced in warm regions or often only in 
the tropics include the diverse types of Citrus fruits such as various 
Oranges, Lemon, Grapefruit, Lime, Citron, Tangerine, Kumquat, 
and products of their hybridization, Bananas of various kinds, 
Custard-apples and their allies, Dates, Durian, Figs of various kinds, 
Guavas and their allies, Granadillas of various kinds, Jujube, Litchi, 
Loquat, Mamey, Mangoes and their allies, Mangosteen, Olive, 
Papaya, various Persimmons, Pineapple, Pomegranate, Sapodilla and 


its allies, Tamarind, and many more. Jams and other preserves 
are chierlv made from fruits and sugar, often with the addition of 
some flavouring, preservative, and or stiffening principle. 

The above outline, which serves to indicate the range and variety 
as well as importance of plant foods for Man, is exclusive of beverages 
and such adjuncts as spices and other flavourings, which will be 
dealt with in the next section. The dependence, on green plants, 
of practicallv all other forms of life either directly or indirectly for 
food, goes of course for the animals which are used extensively by 
Man for his own sustenance, and so we should here recall this 
further dependence of Man on the natural vegetation or cropping- 
possibilities of each region. Many of the above-mentioned cereal 
Grasses, food Legumes, and ' vegetables ' such as Kales and Turnips 
are grown partly for domestic animal fodder, as are, in addition, 
numerous Grasses such as Timothy, Sudan-grass, Johnson-grass, 
Orchard-grass, Redtop, Bluegrass, etc. Furthermore there is an 
important class of non-grassy forage crops which, like some of the 
above-mentioned cereal and other plants when green, are often used 
for silage. These include Mangel-wurzels and such leguminous 
plants as Alfalfa, various Clovers and Vetches, Kudzu, and 
Lespedezas, in addition to Peanut, Soybean, Cowpea, and others 
among those already mentioned in different connections. These 
and other Legumes, together with various Grasses, largely make up 
hay. From the utilitarian point of view, forage plants may be 
looked upon as a means of turning plant carbohydrate and protein 
into meat and dairy products. 

A number of usually minor foods are afforded by the lower plants. 
Thus the use of Mushrooms, Truffles, Morels and other Fungi is 
ancient and familiar, Mushrooms in particular being widely cultivated. 
Food Yeast is another important fungal product. Although the use 
of Lichens for human food has largely died out except in times of 
severe shortage in northern regions, they are still important in the 
feeding of the Reindeer on which whole cultures of boreal peoples 
are largely centred. More widely used for human food nowadays 
are marine Algae, which in the Orient and some Pacific Islands 
constitute a major article of diet, many being cultivated, especially 
in Japan. In Europe and North America the only Algae that are 
at all extensively used in food are Carrageen or Irish-moss, Dulse, 
Murlins, those that give agar and some other products, and various 
Lavers. But extensive research is in progress on the possibility of 
using cultures of freshwater Algae, such as Chlorella, for food. 

260 introduction to plant geography [chap. 

Beverages and Flavours 

The major non-alcoholic beverages, although not as vitally essential 
as the major foods, are yet so familiar that their significance is in 
no need of explanation. They are tea, which is used by fully one- 
half of the population of the world, coffee, which is used by almost 
as many people, and cocoa — the ' beans ' affording this last also 
yield chocolate and cocoa butter. Other beverages include mate or 
Paraguay-tea, obtained from the leaves of various species of Holly ; 
guarana, from the seeds of an Amazonian climber ; cola, obtained 
by powdering Cola seeds ; khat, a tea-like drink of northeastern 
Africa, and cassine, a rather similar beverage of North America ; 
also yoco, which is made from the bark of a South American tree. 
All these beverages contain some cafTeine and consequently have a 
stimulatory and refreshing effect ; and numerous others are, or once 
were, widely prepared from parts of various plants. 

Other non-alcoholic beverages are the so-called ' soft drinks ' 
which include a vast array of preparations that tend to rise and fall 
in popularity — more, perhaps, with the amount of advertising 
lavished by their producers than with their inherent value, though 
most contain a fair amount of sugar and so are a source of energv. 
Many contain plant flavourings, etc., such as ginger, sarsaparilla, 
malted Barley, wintergreen, cola, or fruit juices — the last constitut- 
ing many popular and valuable drinks. 

Of alcoholic beverages there are two main groups : the fermented 
ones in which the alcohol is formed by the fermentation of sugar, 
and the distilled ones obtained by distillation of some alcoholic 
liquor. The sugar is either present naturally, as in most fruit juices, 
or is formed by transformation of starch — for example in cereals or 
potatoes. Wines, of which there are almost endless types of varying 
delegability and to suit different palates and pockets, are the oldest 
and most important of the fermented beverages. They are mostly 
formed by fermentation of sugar in the juice of grapes through the 
activity of wild Yeasts present on the skins of the fruit, though a wide 
range of other plants and their products may be similarly employed. 
The agreeable aroma and flavour are due to the presence of various 
aromatic substances, though the characteristic ' bouquet ' develops 
only after some years or even decades of ageing. 

Beer, ale, and the like make up the other most important group 
of fermented beverages. In their production, cereal starch (usually 
in Barley) is transformed into sugar in the process of malting, and 


this sugar in turn is dissolved out and the resulting product flavoured 
bv boiling with Hops, after which Yeast is added to bring about 
alcoholic fermentation in the main process of brewing. In addition 
there are numerous relatively minor fermented alcoholic beverages 
including cider, made from the juice of Apples ; perry, from Pear 
juice ; mead, from honey and water ; sake, from Rice ; Palm wine, 
from the juice of Palm inflorescences ; chicha, from Maize ; and 
various so-called beers made from infusions of various roots and 
barks, with the addition of sugar and yeast. Furthermore, acetic 
acid fermentation by Bacteria leads to the formation of vinegar, 
another widely used product. 

The chief distilled alcoholic beverages (' spirits ') are made by 
successive distillations of fermented mashes or wines. Thus whisky 
is obtained from malted or unmalted cereals or potatoes and, after 
distillation, has to be aged to eradicate unpalatable principles. 
Vodka on the other hand is bottled directly after distillation. Brandy 
is distilled from wine, or, in the case of fruit brandies, from fermented 
fruit juices. Good gins are obtained from a mixed mash of barley- 
malt and rye, the flavour being due to added oil of juniper or other 
aromatic essential oils. The numerous liqueurs and cordials con- 
sist mainly of sugar and alcohol or spirits flavoured with various 
essential oils, being often blended according to some secret formula. 

Spices, condiments, and other food adjuncts are almost innumer- 
able, so only the most important can be mentioned here. Man's 
craving for spices has done much to change the course of history 
and affect international and inter-racial relations. The value of 
spices and condiments lies in their ability to increase the attractive- 
ness of food, etc., usually owing to the presence of essential oils. 
Besides their use as food adjuncts, spices are employed in various 
industries, including perfumery, drug and soap manufacture, dyeing, 
and in the arts. The vast majority are still obtained from the tropics, 
chiefly from Asia. 

Most of these flavouring materials originate in seeds and fruits. 
They include allspice or pimento, obtained from a small tree of 
tropical America ; capsicum or red pepper (including chilis, paprikas, 
and sweet peppers), from several plants now widely cultivated ; 
black and white and some other peppers, from weak climbing or 
trailing shrubs that are widely cultivated in the tropics ; cardamon 
and grains of paradise ; the various mustards (black, white, and 
Indian), from allies of the Cabbages growing often in cool climates ; 
nutmeg and mace, from the Nutmeg tree which is now cultivated 


principally in the East and West Indies ; anise and star anise, cara- 
way, coriander, dill, fennel, vanilla and its substitute tonka beans, 
and many others. 

Spices obtained from flowers or flower-buds include cloves, 
capers, and saffron, while from leaves are obtained such familiar 
ones as peppermint, balm, basil, marjoram, sage, the savouries, 
spearmint or mint, bay, thyme, lemon thyme, parsley, wintergreen, 
tansy, and tarragon, most of which can be (and commonly are) 
grown in gardens in temperate regions. Spices or flavourings 
obtained from barks include cinnamon and cassia, from various 
species of Cinnamon trees grown chiefly in southeastern Asia, and 
sassafras, from the familiar North American tree of that name. 
Important spices obtained from roots and rhizomes are angelica 
(other parts of the Angelica plant are also used), now cultivated 
chiefly in Germany ; ginger, from the Ginger plant which is widely 
cultivated in the tropics ; horse-radish, widely grown in temperate 
regions ; sarsaparilla, from several tropical Catbriars ; and turmeric, 
cultivated in various tropical regions and employed to colour as well 
as flavour curries, etc. 


The history of the medicinal use of plants is long and intricate, 
being largely bound up with the beliefs of primitive peoples — for 
example, that disease was due to the presence of evil spirits in the 
body, which could be driven out by the use of unpleasant substances. 
After the Dark Ages came the herbalists with their compilations of 
what was known or supposed about the medicinal value and folk- 
lore of plants, and the ' doctrine of signatures ' according to which 
plants were supposed to possess some sign indicating the use for 
which they were intended. Thus the Maidenhair Fern, so called 
from the black hair-like ' stalks ' of the leaflets, was considered a 
specific for baldness, and plants with heart-shaped leaves were 
believed to be valuable for use against heart ailments. Plants were 
believed to have been placed in the world for Man's use, and to 
have clear indications of their particular usefulness provided by 
the Almighty. 

From such crude beginnings developed modern pharmacognosy, 
which is concerned with the knowledge and commerce of crude 
drugs and their sources, and pharmacology, the study of the action 
of drugs. The medicinal value of drugs is due to the presence in 


them of special substances having a particular physiological action 
on the human body : commonly such substances are alkaloids, some 
of which are powerful poisons if administered unwisely, while others 
are dangerously habit-forming. Yet in small quantities skilfully 
administered, even the most poisonous or dangerous drugs can be 
of value to human health and well-being. 

Throughout the world there are used for medicinal purposes some 
thousands of different plants or plant products — many of them only 
locallv bv savage peoples. The use of others has been rendered 
obsolete bv synthesis of their active principles. Some are widely 
cultivated, but many more are gathered chiefly or entirely in the 
wild state and are still important commercially. A few of the most 
significant, classified according to the part of the plant from which 
they come, are the following: 

Obtained from fruits and seeds : chaulmoogra oil, from a south- 
east Asian tree, containing principles effective in the treatment of 
leprosy ; colocvnth, from a widespread perennial vine now cultivated 
in the Mediterranean region, serving as a powerful purgative, as 
does also croton oil, obtained from a shrub or small tree of south- 
eastern Asia ; nux vomica, obtained from a tree ranging from India 
to Australia, used as a stimulant in small quantities as it contains 
strychnine ; strychnine itself, obtained from the same and allied 
plants ; opium, obtained as an exudation from the injured fruits of 
the widely cultivated Opium Poppy, employed to relieve pain but 
flagrantly misused as a narcotic in the Orient ; psyllium, from 
Plantains, used chiefly as a laxative ; strophanthus, from two 
African lianes, used as a heart stimulant ; and wormseed, a native 
of the warm parts of the New World, used in the treatment of 
hookworm infections. 

From flowers are obtained chamomile, which is rather widely 
cultivated and used for a variety of purposes ; hops, extensively 
cultivated in temperate regions and used for their sedative and tonic 
properties as well as in brewing ; and santonin, one of the best 
remedies for intestinal worms. 

The vegetative parts of plants contribute numerous drugs. From 
leaves are derived, for example, aloe, from African and American 
Aloes, used as purgatives ; belladonna, from the Deadly Nightshade 
of Europe, etc., but now extensively cultivated, used for the local 
relief of pain and for a variety of other commendable purposes ; 
cocaine, from the leaves of the South American Coca shrub that is 
now extensively cultivated in the tropics, used chiefly as a local 


anaesthetic ; digitalin, from the leaves of the European Foxglove, 
almost indispensable in the treatment of heart ailments ; eucalyptus, 
from various Blue Gum-trees of Australia but now widely cultivated 
elsewhere, extensively used in medicine, for example in the treat- 
ment of nose and throat disorders ; hamamelis, from the North 
American Witch-hazel, used as an astringent ; henbane, from the 
widespread weedy herb of that name, used as a sedative and hypnotic ; 
stramonium, from the widespread and weedy Thorn-apple, used as 
a narcotic and in the treatment of asthma ; and many others. 

From stems, etc., come ephedrine, from Asiatic Ephedras, used 
in medical treatment e.g. of colds and hay-fever ; guaiacum, used 
as a stimulant and laxative ; and quassia, used as a tonic and in 
the treatment of malaria. From barks are obtained cascara, from 
the American Western Buckthorn, used as a tonic and laxative ; 
curare, from a variety of South American plants, a very powerful 
poison used also in medicine and surgery ; slippery elm, made from 
the familiar North American tree, useful for its soothing effect ; 
and, above all, quinine, the great anti-malarial drug, obtained from 
several allied trees native to South America and now cultivated 
especially in southern Asia. 

From roots and other underground parts come an important series 
of drugs including aconite, from the Eurasian Monkshood, used 
particularly to relieve pain ; colchicum, from the Meadow Saffron, 
whose active principle is colchicine, used in the treatment of 
rheumatism and gout and also important in the plant sciences as it 
produces doubling of chromosomes ; goldenseal, from the North 
American plant of that name, used as a tonic and in the treatment 
of catarrh ; ipecac, from forest-floor plants of tropical America, 
almost indispensable in the treatment of amoebic dysentery and 
pyorrhoea ; liquorice, from the Liquorice plant now much cultivated 
in Eurasia, used as a flavouring, etc. ; squills and senega, used as 
expectorants and stimulants ; ginseng, considered a virtual cure-all 
in the Orient ; and valerian, from Garden Valerian, used to relieve 
nervous afflictions. 

From various parts particularly of the Camphor Tree come 
camphor and safrole, which have a wide variety of industrial and 
medicinal uses. Among Pteridophvtes, the tiny spores of some 
Club-mosses are variously used for covering pills as well as in 
industry and even warfare, while aspidium, obtained from certain 
Ferns of the north-temperate regions, has long been employed to 
expel Tapeworms. 


The lower plants come notably into their own as sources of drugs : 
penicillin, streptomycin, aureomycin, Chloromycetin, terramycin, 
and others recently developed from Fungi and their allies are being 
found extremely valuable in the treatment of some of the most 
severe diseases. Ergot, produced from the common fungal disease 
of cereals bearing the same name, has long been known and is used 
in the treatment of haemorrhages. Agar, widely obtained from a 
number of Red Algae, is employed medicinally to prevent constipa- 
tion, as well as industrially in food, paper, and cosmetic manufacture, 
and as a culture medium for Fungi and Bacteria. Algin, a product 
of the larger Kelps, is used in a variety of cosmetics, foods, drugs, 
and as a sizing for paper; these Brown Algae are also important 
as sources of iodine and potash. 

Fatty Oils and Waxes 
Fatty or fixed oils are those which, unlike the essential oils (see 
pp. 274-5), d° not easily evaporate, and so cannot be distilled without 
becoming decomposed. Those which are liquid ' oils ' at ordinary 
temperatures become solid ' fats ' on cooling, even as fats become 
oils on sufficient warming. Like animal fats, they consist of glycerin 
in combination with a fatty acid, and form soaps when boiled with 
alkalis. Fatty oils (as they may in general be termed) are produced 
in considerable quantities by many different plants, often being 
stored in seeds for use in germination. Of them four main classes 
may be recognized, as follows : 

1. Drying oils, which on exposure dry into thin elastic films 
and are of great importance in the paint and varnish industries. 
Examples include linseed oil, obtained from Flax, and its substitute 
tung oil, obtained from two Chinese trees ; soybean oil, which is 
widely used in human foods as well as industry, etc. ; and such 
oils as perilla, walnut, Niger seed, hempseed, poppy, and safflower. 

2. Semi-drying oils, which form a soft film only after long exposure, 
examples being cottonseed oil, obtained from Cotton, and used for 
human food, animal fodder, fuel, and in a variety of industries ; 
sunflower oil, obtained from the common Sunflower, used for the 
same purposes as cottonseed oil and also in the paint, varnish, and 
soap industries ; and such oils as corn, rape, and camelina. 

3. Non-drying oils, which remain liquid at ordinary temperatures. 
These include olive oil, obtained from the Olive and used principally 
for food and in medicine (though inferior grades are employed in 
soap-making and as lubricants) ; its widely-used substitute, peanut 


oil, obtained from the Peanut ; and castor oil, obtained from the 
Castor-oil plant and formerly used chiefly in medicine but now much 
more widely in industry. 

4. Vegetable fats or tallows, which are more or less solid at ordinary 
temperatures. These include coconut oil, obtained from the ' meat ' 
of the Coconut, and widely used for making margarines, candv- 
bars and other sweets, soaps, cosmetics, stock-feed, and as an 
illuminant ; palm oil, obtained from the African Oil Palm and used 
for some of the same purposes as coconut oil ; palm-kernel oil, the 
Brazilian palm oils, cocoa butter, and many others. 

Waxes are chemically allied to fats but tend to be harder. They 
usually occur as coverings of leaves, etc., and help to prevent too 
great a loss of water by transpiration. Among those of value to 
mankind are carnauba wax, obtained from a South American Palm, 
widely used in the manufacture of candles, soaps, paints, varnishes, 
ointments, and many other products ; cauassu wax, which can be 
used for much the same purposes ; and the American candelilla, 
myrtle, and jojoba waxes. 

Lather-forming products of the cultivated Soapwort, of the 
South American Soapbark tree, of the Soapberry, and of California 
Soaproot, are used commercially as soap substitutes — as are many 
others more locally. 

Smoking and Chewing Materials 

Of these ' fumitories and masticatories ' the most universally 
employed is tobacco, obtained principally from a tropical American 
species that is now very widely cultivated, but to some extent also 
from a North American species which was extensively grown by 
the Indians before the time of Columbus. Important chewing 
materials are betel, obtained from the widely cultivated Betelnut 
Palm and said to be used by over 400,000,000 people ; cola nuts, 
from the West African Cola tree which has now been widely intro- 
duced elsewhere, whose use results in slight stimulation and tem- 
porary increase in physical capacity ; and, in addition, chewing gum 
and spruce gum. 

In contrast to the above more or less harmless principles, the 
true narcotics contain powerful alkaloids that make them gravely 
detrimental to human health if used habitually, although they are 
valuable in medicine in exceedingly small amounts. Among the 
chief are opium and cocaine, obtained as already mentioned above 


under drugs, and constituting important relievers of pain. The 
eating or smoking of opium, which at first produces alluring dreams 
and pleasurable visions, may become an uncontrollable addiction 
leading to delirium and death. The chewing of the leaves of the 
Coca shrub, containing cocaine, gives resistance to fatigue and 
hunger and at the same time a feeling of exaltation; but if habitual, 
it may in time lead to severe physical deterioration and even death. 
Other important narcotics are cannabis, obtained from the Hemp 
plant and widely used in medicine to relieve pain as well as in the 
treatment of nervous disorders ; fly agaric, obtained from the 
poisonous Fungus of that name, which when chewed or added to 
beverages has an intoxicating effect involving hallucinations and 
finally unconsciousness ; peyote (mescal buttons), from an American 
Cactus, mostly chewed for the feeling it produces of well-being 
accompanied by hypnotic trances ; products from Thorn-apples and 
Henbane, which when smoked or eaten produce excitations, illusions, 
and sometimes fanatical acts ; and the Oceanian kavakava, whose 
use as a beverage has a sedative and soporific action, bringing about 
pleasant sensations. Cannabis consumption as hashish, marijuana, 
etc., causes states of ecstasy and stupefaction and may have serious 
results, as when it leads to fanatical acts by addicts. 

Structural and Sheltering Materials 

It can scarcely be questioned that wood is nowadays, and from 
before the dawn of history has been, the most generally important 
of all structural materials. Its uses in the construction of human 
habitations and shelters, furniture and utensils, vehicles and boats, 
tools and all manner of fittings and fencings, etc., are too universally 
familiar to require detailed enumeration. If we reflect that until 
a century ago ships were made almost entirely of wood, without 
which the exploration and colonization and general development of 
most of the world would accordingly have been impossible, we have 
one impressive indication of its immense significance in human 
affairs ; and to this day wood is the most widely used commodity, 
apart from foodstuffs and, perhaps, clothing materials. It is also 
highly versatile as a raw material (and one of the few that can per- 
petually be renewed) for conversion into products as varied as paper 
and textiles, soap and lubricants, stock-feed and motor fuel, plastics and 
disinfectants, explosives and preservatives (to mention only a few). 

Quite apart from questions introduced by their local availability, 


different woods vary markedly in mechanical and allied properties 
and, consequently, in usages. Important features are general 
strength, hardness, stiffness, toughness, fineness of grain, cleavability, 
density, moisture content, the commonness of defects, and suscepti- 
bility to insect damage and to decay. Expression of many of these 
depends upon the age of the tree from which the wood was cut 
and the treatment it has received since cutting, and all of them 
naturally vary with the kind of tree involved. It is said that the 
annual consumption of wood in the world amounts to some sixty 
thousand million cubic feet, of which nearly half is used in North 

The two main types of timber are softwood, obtained from 
coniferous trees such as Spruces, Pines, and Larches, and hardwood, 
from Angiospermous trees such as Oaks, Maples, and Mahoganies. 
Besides its use as a popular fuel and employment as a raw material 
for conversion into diverse products whose origin is often unrecogniz- 
able, wood is used, as such, in the form of structural timber for 
buildings and bridges, as boarding and flooring, girders and rafters, 
pit-props and railway ' sleepers ', poles and posts, piling and cooper- 
age, veneers and plywood, shingles and woodenware, parts or the 
whole of ships and boats, furniture and vehicles, boxes and crates, 
fences and hoardings, and for innumerable other purposes of which 
some have already been mentioned. 

Among specific sheltering materials other than wood, the leaves 
of such woody plants as Palms, and herbaceous materials such as 
straw, are widely used for shelters and thatches in different parts 
of the world. Also extensively employed are various soft or com- 
minuted plant products for insulating and packaging, familiar 
examples being Bog-mosses and hay for insulation, and sawdust and 
shavings for packaging. 

Industrial Uses and Extractives 

Many of the items discussed under other captions, such as fuels, 
fats, and fibres, might be included here, but the chief classes to be 
considered in this section are those whose chemurgical treatment or 
processing makes them important sources of industrial derivatives 
— namely, sugars, starches, cellulose, and some distillation products. 
To take this last item first, the destructive distillation of wood- 
waste produces such valuable materials as charcoal, tar, oil, turpen- 
tine, wood alcohol, acetic acid, and wood gas. 


Sugar, as we have already seen, is obtained principally from the 
Sugar Cane and the Sugar Beet, and though primarily used as food 
it has also become an extremely important industrial chemical, with 
thousands of different derivatives. About 35,000,000 tons are pro- 
duced annually. When speaking of sugar we normally mean sucrose, 
which is by far the most important and the one usually stored, 
though other sugars have their places and uses. Further important 
sources of sucrose are Maize, Sugar Maple, Sorghum, certain 
Palms, and honey. 

Starches constitute the chief type of food-reserve for most green 
plants, being stored in the cells in the form of minute grains. 
Starches are chemically complex but, being easily convertible into 
sugars, are vastly important as human foods, the chief sources being 
potatoes, various cereals, arrowroot, cassava, and sago. They are 
also widely used in industry, for example in laundry and textile 
work, in sizing, and as sources of glucose, dextrin, industrial alcohol, 
and explosives. 

Still more complex is cellulose, the chief constituent of the cell- 
walls of most plants, which yields numerous textile fibres both 
natural and artificial. The chief sources nowadays are cotton, which 
is almost pure cellulose, and wood, which by chemical and mechanical 
treatment is made to yield pure cellulose. This may be made into 
fibres or plastics, or transformed into wood sugar, which in turn 
may be made to nourish Yeast or yield alcohol and thus become 
available for food or industrial use. In addition there is hemi- 
cellulose, which forms the so-called ' vegetable ivory ' — obtained 
from certain tropical Palms and useful as a substitute for ivory in 
the manufacture of buttons and other small, hard objects. 

While paper can be made from most fibrous materials, the chief 
commercial sources are wood fibres, cotton, and linen. The last 
two, formerly the main source of paper, still yield the finest grades, 
but wood fibres nowadays make up the vast bulk. They are 
obtained from a wide range of trees of which various Spruces, 
Pines, Hemlocks, and Poplars are among the most important, while 
sawmill waste is increasingly used. Other raw materials for paper- 
making include papyrus, esparto, straw, mulberry, and various textile 
fibres. By special processes the wood or other raw material is 
pulped, after which a series of operations, including the addition 
of rosin or other ' sizing ' of plant origin, lead to its manufacture 
into one or another of the almost innumerable types and grades of 
paper. The coarser materials are often made into cardboard. 



Cellulose is soluble in various solvents such as concentrated nitric 
acid, and this has led to the development from it of ' plastics ' and 
unbreakable glasses and many other important products including 
guncotton, cordite, collodion, celluloid, cellulose acetate, viscoses 
such as Cellophane, and various varnishes and fabrics including cloth 
and leather substitutes. Photographic film is made chiefly from 
cellulose nitrate or acetate coated with gelatin, while further breaking 
down of cellulose yields various sugars which in turn yield alcohol 
and, with appropriate treatment, Torula Yeast or other foods. Alto- 
gether cellulose products are countless in number and of great value 
and usefulness ; moreover, being formed often from forest waste, 
they seem endlessly replaceable. 

Clothing Materials and Other Fibres 

Clothes, one of Man's primary requisites, are made largely from 
plant fibres or materials obtained from animals which are dependent 
on plants for food. In addition, fibres and fabrics are used by 
Man in innumerable other ways. Indeed, fibre-yielding plants are 
probably second only to food plants in their influence on civilization 
and their general usefulness to Man. But although there are many 
hundreds of fibre-yielding plants known, only a few are of com- 
mercial importance. 

From the point of view of their utilization, fibres produced by 
plants or from plant materials may conveniently be classified into 
seven groups : (1) Paper-making fibres, which were discussed in 
the last section dealt with above. (2) Artificial fibres, whose pro- 
duction is nowadays a great and expanding industry. The main 
raw materials are (a) cellulose, derived from wood pulp or cotton 
linters, whence are made the various so-called ' rayons ', and (b) 
soft coal, whence are produced the various nylons and related 
materials which between them have now some hundreds of important 
uses. (3) Textile fibres, used as fabrics, netting, and cordage. 
These include cotton of various kinds, flax, hemp, jute, ramie, 
Manila hemp, sisal, coir, and many others. (4) Brush fibres, which 
are tough and stiff, the chief being the piassavas and their allies, 
obtained from certain Palms, and Grasses such as Broomcorn, 
Broomroot, and Spartina. (5) Plaiting and tough-weaving fibres, 
employed for making straw hats, baskets, chair-seats, matting, 
wickerwork, etc., for which the stems of various Palms, Grasses, and 
grass-like Sedges and Rushes arc used. (6) Filling fibres, used for 


upholstery, stiffening, packaging, and caulking, the outstanding 
example being kapok (see p. 235) and its various substitutes, which 
include the silky hairs on the seeds of the familiar Milkweeds. 
There is also Spanish-moss, an excellent substitute for horsehair. 
(7) Natural fabrics, etc., consisting of tough interlacing fibres that 
can be extracted from bark in layers or sheets and used as a sub- 
stitute for cloth. Examples include the Polynesian and Oriental 
tapa cloth and the Jamaican lace-bark, as well as such fibrous pro- 
ducts as the so-called vegetable sponges or luffas which are used 
for making hats, for scouring, for filtering, and as substitutes for 
bath sponges and body scrapers. 

Fuels (including Fossil, etc.) 

Fuel, as a source particularly of heat, light, and power, is one of 
the greatest necessities of human life, and in general consists of 
plants or plant products whether modern or belonging to earlier 
epochs. A few of the main groups of plant materials that are widely 
used as fuels may be outlined : (1) wood, probably still used more 
for fuel than for any other purpose, certain hardwoods being in 
general better than other types, but almost all woods making useful 
fuels when dry ; (2) vegetable oils, used principally in this connection 
for illumination and for powering Diesel engines ; (3) peat, con- 
sisting of compacted deposits of partially decomposed vegetable 
matter, which is widely used for heating and cooking in northern 
lands especially where wood is scarce ; (4) manure, which is the 
almost universal fuel of hundreds of millions of people in southern 
Asia ; (5) coal, the compressed and fossilized remains of plants that 
lived in much earlier geological epochs and are now largely decom- 
posed and converted into carbon, being a valuable and widely used 
source of fuel and power and also of gases which are employed 
particularly for heating and illuminating ; (6) coke, which is left 
when coal-gas is driven off from coal, and is nearly pure carbon, 
forming an excellent fuel which burns without appreciable smoke 
or flame ; (7) charcoal, which bears a similar relationship to wood, 
and is the chief domestic fuel in many tropical countries ; (8) saw- 
dust, etc., used principally in the form of briquettes ; and (9) 
petroleum, whose familiar products of distillation include the all- 
important gasoline (petrol) as a source of power, and paraffin and 
kerosene as sources particularly of heat and light. Although no 
trace of plant structure remains in petroleum, it is generally supposed 


to have had its origin in minute primitive forms of plant life that 
flourished in much earlier geological times. 

Latexes and Exudates 

Of the products obtained from the milky juice (latex) of various 
plants, rubber is by far the most important. In spite of the con- 
siderable employment of synthetic forms, over a million tons of 
natural rubber are used annually, most being produced in south- 
eastern Asia. Over three-quarters of the crude rubber consumed 
goes into tyres and inner tubes, while other important uses are in 
footwear, packaging, waterproof clothing, road construction, tubing 
and belting, electrical insulation, etc. Rubber is produced princip- 
ally from various tropical woody plants of the Spurge, Mulberry, 
and Periwinkle families, some of which are now cultivated. Besides 
wild and plantation rubbers produced from the Para (Hevea) Rubber 
tree, which is by far the most important source of rubber, there are 
such other natural types, produced from different sources, as Assam, 
castilla, ceara, guayule, and even dandelion rubbers, the last being 
cultivable in the cool-temperate belt. 

Other latex products include the non-elastic gutta-percha and 
balata, obtained from tropical trees and used for insulation, piping, 
golf-balls, telephone receivers, and many other purposes, and chicle, 
obtained from the tropical American Sapodilla tree, which is the 
basis of the chewing-gum industry. 

The gums which exude from plant stems either naturally or in 
response to wounding, and the resins which are usually secreted in 
definite cavities or passages, may conveniently be classed together 
as exudates. Gums are used principally as adhesives, as sizing for 
paper, in medicine and polishing, in cosmetics and ice-cream, in 
printing and finishing textiles, as a glazing for paintings, and in the 
confectionery and paint industries. The chief commercial varieties 
are gum arabic, obtained from certain Acacias of arid northern 
Africa, gum tragacanth, from certain Milk-vetches of arid south- 
western Eurasia, and its substitute karaya, obtained from a tree in 
India whence several million pounds are exported annually. The 
related pectins are widely used to make foods jell, in pharmaceuticals 
and cosmetics, in sizing and adhesives, fibres, films, and other 
preparations. They come chiefly from citrous and apple wastes, 
but many other fruits, etc., afford potential sources. 

Resins are more various and tend to be still more important than 


gums, though usually tapping is necessary to obtain them in com- 
mercial quantities, or they may be collected in the fossil state. For 
the most part they are forest products. Some of the more valuable, 
with their uses, are (1) the various copals, utilized as varnishes and 
in making paints and linoleum; (2) amber, a fossil resin particularly 
from an extinct species of Pine, which is used for beads, ornamental 
carvings, and mouth-pieces of pipes, etc.; (3) damars, used princip- 
ally in varnishes; (4) lacquer, a natural varnish which hardens on 
exposure to air and affords remarkable protection even against acids 
and alkalis; (5) shellac, excreted by a particular insect on twigs of 
certain trees on which it feeds, widely used in insulation and decora- 
tion and for making varnishes, sealing-wax, size, drawing inks, 
gramophone records, and many other products ; and (6) turpen- 
tines, chiefly obtained by tapping coniferous trees and yielding on 
distillation oil of turpentine and rosin. Oil of turpentine is of major 
importance in the paint and varnish industry as a solvent and thinning 
agent, and in chemical manufacture, while rosin is the chief sizing 
material for paper and is also used in many manufactures as well as 
in greases and lubricants. 

Other noteworthy products in this general category include Canada 
balsam, used in mounting microscope slides and as a cement for 
lenses ; spruce gum, used as a masticatory ; Venetian turpentine, 
used in varnishes and veterinary work ; and various balsams, used 
in medicine, adhesives, soaps, lotions, and cosmetics, as well as for 
the flavouring of foods and as fixatives in the perfume industry. 
There are also such products as ammoniacum, used in medicine 
and perfumery ; asafoetida, widely used in medical treatment ; 
copaiba, used for making varnishes and lacquers, as a fixative of 
perfumes in soap, and in photography and medicine ; elemi, used 
in various artistic, cosmetic, and medical operations ; frankincense 
used in incense, cosmetics, and fumigation ; and myrrh, used for 
cosmetic and medicinal purposes as well as in incense and embalming. 

Tanning and Dyeing Materials 

Tanning involves the reaction of strongly astringent tannins with 
such proteins as are present in animal skins, thus forming the strong 
and resistant, flexible commodity we know as leather. Although 
tannins are very widespread in plants, relatively few species are 
known to contain a sufficient proportion to be of commercial import- 
ance, and these are in great demand. The sources occur mostly 


in the wild state and include the woods 'of Quebracho and Sweet 
Chestnut, the leaves of Sumac and Gambier, such fruit products as 
myrobalan, tara, and valonia (from the Turkish Oak), root materials 
from Tanner's Dock and Palmetto, and barks of Hemlock, various 
Oaks, Mangroves, Eucalypts, Wattles (Acacias), Larches, Spruces, 
Birches, and Willows. However, the tannin content in the last four 
instances is too low to warrant their general use. Tannin-inks are 
the most important inks at the present time, their tannin being 
derived largely from the insect galls formed in great abundance on 
the twigs of the Aleppo Oak ; to these galls or an extract made from 
them are added an iron salt, an agglutinant such as gum arabic, and 
a colouring matter such as logwood {see below). Other inks, too, 
are made substantially from plant products. 

Of natural dyes and stains obtainable from plants there is a vast 
array involving almost all colours, though latterly most have been 
supplanted at least in part by the synthetic or aniline dyes obtained 
from coal-tar products. Such ' artificial ' dyes tend to be brighter, 
cheaper, and more lasting, and with them only a few vegetable dyes 
compete nowadays. These vegetable dyes are especially useful in 
dyeing fabrics, but are also employed to colour a wide range of 
other familiar products. Some of the more important of these 
natural colouring matters, grouped according to the plant part from 
which they come, are : from seeds and fruits — annatto, Persian berries, 
and sap green ; from flowers — safflower and saffron ; from barks 
— quercitron, lokao, and gamboge (an exuded gum resin) ; from 
leaves — indigo, henna, woad, and chlorophyll (which is harmless 
and consequently used in foods and toothpastes) ; from woods — 
logwood or haematoxylin (of which many thousands of tons are 
used annually to give colours ranging from reds to purples and 
black), fustic, cutch, osage orange, sappanwood, and brazilwood ; 
and from roots and tubers — alkanna, madder, and turmeric. Lichens 
yield some fine dyes, among which archil and litmus still find 
extensive use. 

Essential Oils and Scents (Perfumes) 

Very different from the fatty oils already considered are the 
so-called essential oils which have a pleasant taste and strong aromatic 
odour, easily volatilizing in air. They are complex in chemical 
nature but tend to be readily removed — by distillation, expression, 
or solvent-action — from the many and various plants that produce 


them. Their main uses are for scenting, flavouring, or medicinal 
purposes — for example in the manufacture of perfumes, soaps, and 
other toilet preparations, in cooking and the production of all 
manner of foods and beverages, and for therapeutic, antiseptic, and 
bactericidal purposes. Other uses are as clearing agents and solvents, 
as insecticides and deodorants, and in such diverse products as 
printer's ink and toothpaste, library adhesives and chewing-gum, 
shoe-polish and tobacco. 

We have already dealt with the flavouring and medicinal sides, 
and noted the industrial uses, of various essential oils, and shall be 
concerned with insecticides, etc., in the next section ; here we 
must mention some of the more important ' essentials ' used in 
perfumes (or scents, as they are called in Britain). These often 
highly-priced products commonly consist of blends of the essential 
oil or oils in alcohol, usually with a less volatile fixative. Examples 
are rose oil or otto (attar) of roses, obtained principally from flowers 
of the Damask Rose which are distilled without delay after being 
picked in the early morning just as they are opening ; orange 
blossom oil (neroli), formerly obtained from citrous plants grown 
for the purpose but nowadays often synthesized and used in cos- 
metics, etc. ; lemon-grass oil, from the leaves of a particular Grass 
and used in cosmetics and medicine ; oil of citronella, from another 
member of the same genus and used in cheap perfumes and as a 
deodorant and insect repellent ; geranium, distilled from the leaves 
of various species of ' Pot Geraniums ' (Pelargoniums) and widely 
used in making perfumes and soaps ; ylang-ylang, from the flowers 
of an Asiatic tree and now said to be present in almost every perfume ; 
cassie, from one of the Acacias ; cedarwood oil, from the Eastern 
Red Cedar ; bergamot, from a type of Orange ; bay rum, from a 
West Indian tree ; calamus, from the Sweet-flag ; camphor (in 
spite of its solid form) ; lavender and rosemary which are used 
extensively in eau-de-cologne and soaps ; and very many others. 
Although numerous synthetic products are now available, most of 
the best perfumes are still of botanical origin and in increasing 
demand, the annual gathering of flowers for this purpose alone being 
said to exceed 10,000,000,000 lb. (over 4J thousand million kilos). 

Insecticides and Herbicides 

Although more than twelve hundred species of plants have been 
reported to have some insecticidal or at least insect-repellent 


properties, the vast majority are of little importance and even the 
best tend to be overshadowed nowadays by such synthetic insecticides 
as DDT. Nevertheless some plant products continue to be used 
very extensively and even increasingly to combat insects and other 
vermin, and accordingly to be of great service to Man. The three 
most important are : (1) nicotine, extracted from leaves of Tobacco 
plants and used as a spray that is lethal to some of the worst insect 
pests ; (2) rotenone, obtained from the roots of various tropical 
trees or shrubs belonging to the Pea family, and long used as a fish- 
poison as well as, latterly, in powdered form or spray to kill various 
insect pests of crops and livestock ; and (3) pyrethrum or insect 
flowers — likewise used as dusts or sprays that quickly paralyse 
insects including pests afflicting Man, and obtained from the flower- 
heads of certain members of the Daisy family that are widely cul- 
tivated for the purpose. 

There are also such repellents, etc., as camphor, which is obtained 
principally from the Camphor tree but is also synthesized, cedarwood 
oil, the Mexican ' Cockroach plant ', and the Chinese ' Thunder-god 
vine '. Most botanical insecticides are harmless to human beings 
and other warm-blooded animals. However, red squill, obtained 
from the bulbs of a small herb of the Mediterranean region, is an 
important ' raticide ', having little effect on animals other than Rats 
and Mice. 

Whereas the old-fashioned herbicides are usually more or less 
caustic chemical compounds (' weed-killers ') that kill off vegetation 
indiscriminately, investigation of plant growth hormones in recent 
decades has led to the discovery and use of ' hormonal ' herbicides 
that are highly selective in their action in killing some plants while 
leaving others, and animals, unharmed. Outstanding in this respect 
is 2,4-D, which at suitable concentrations kills most dicotyledonous 
plants without affecting most monocotyledonous ones. Thus it 
forms an effective lawn or wheat-field spray, killing most of the 
(dicotyledonous) weeds without injuring the Grass or grain crop. 
Plant growth substances are now synthesized in quantitv and are 
used in various horticultural practices such as the promotion of 
rooting in cuttings. 

Environmental and Ecological 

In spite of the increasing ease and effectiveness of transport in 
the modern world, the availability of this or that plant product in 


a particular place depends to a considerable extent on whether the 
plant from which it comes can be cultivated locally — especially if it 
is bulky and needed in large quantities or in a fresh state, as so many 
foods, etc., are. And quite apart from this direct dependence of 
mankind on the crop and other plant productivity of different areas, 
we get very different environments created by different types of 
vegetation — as was already pointed out in the second paragraph of 
this chapter. Among many other things, trees give shade and 
shelter, and when they are widely aggregated into forests, these 
may regulate climate to a considerable extent, ' damping down ' 
temperature and humidity fluctuations in their shade. In some 
ways, in spite of their transpiration, trees may also help to conserve 
water, for example by preventing run-off and floods, meanwhile 
checking erosion. Forests moreover afford shelter and range for 
livestock and wild animals, and recreation for human beings. 

The widespread use of such ecological devices as sand-binding 
Grasses or other plants and wind-breaking trees and shrubs, is further 
testimony to the value of plant life to mankind. Particularly are 
various surface-binding plants of importance in combating erosion, 
which is one of the world's worst scourges, as is further indicated 
in our concluding two chapters. 

Finally it should be noted here as well as in the next section that, 
in the local environments which Man makes for himself, the import- 
ance of gardens of one sort or another is enormous practically the 
world over. And the supply and use of agricultural, forestral, and 
horticultural implements such as harvesters, saws, and lawnmowers, 
and the general tending of plants, involves vast industries almost 

Aesthetic and Ornamental 

In most landscape views, plants form the chief embellishment, 
and landscapes would suffer greatly without them. Apart from the 
vital needs they satisfy and the material benefits they bestow, plants 
greatly enhance Man's aesthetic appreciation of the world in which 
he lives— both in their natural growth as vegetation, whether 
arborescent or otherwise, and through their cultivation for orna- 
mental purposes. Almost everywhere Man lives, gardens are 
cultivated for recreational and other reasons, and the dustiest city 
streets and drabbest homes are enlivened by greenery and pot or 
cut flowers. 


Floriculture, the branch of horticulture concerned with the com- 
mercial production of flowers, is really a huge industry that is 
important on an almost world-wide scale, but especially in the 
temperate zone. The actual growing is often scientifically regulated 
in many ways, as in greenhouses, and the organization for transport 
and selling is complex and vast. Lawn-growing and landscaping 
are also of considerable importance in urban areas and elsewhere, 
for recreational and ornamental purposes ; their primary function, 
however, is the growing of plants. All this, moreover, involves 
extensive trading in seeds, bulbs, etc., and sometimes in apparatus 
for plant growth — for example in hydroponics, their cultivation 
without soil. 

Microorganisms and Miscellaneous 

We have already mentioned the importance of some micro- 
organisms in affording drugs (such as penicillin) and foods (such 
as Food Yeast), and, towards the end of the last chapter, in causing 
plant diseases. Others are of vast importance in bringing about 
desirable changes — for example the Yeasts in causing fermentation 
of sugars to alcohols, and various Bacteria in effecting further 
fermentation to vinegar as well as in the ' curing ' of tobaccos and 
' ripening ' of cheeses. On the positive side, further valuable 
fermentation, fibre-retting, organic acid and vitamin production, 
hide-dehairing, nitrogen-fixing, sewage-disposal, purification and 
preservation, cooking and silage and all manner of other operations 
are carried out only by or with the aid of microorganisms. On the 
negative side, there are the numerous human and other animal 
diseases which they cause, as well as loss by rotting, putrefaction, 
and general decay. These last activities involve the breakdown of 
complex carbohydrate and other materials to simpler ones and 
ultimately to the raw materials whence they came. Such ' degrada- 
tion ' is essential to keep the world going, for without it the vital 
raw materials such as carbon dioxide would all be used up sooner 
or later and life would come to a standstill. Consequently non- 
green microorganisms, which are almost entirely responsible for the 
breakdown and return of the essentials to general circulation, are 
the world's great scavengers, and, as such, are of fundamental 
importance to all life. 

Nor, among microorganisms, must we overlook the smaller Algae, 
which possess chlorophyll and so are able to build up complex 


foods from simple beginnings. For they afford the ultimate source 
of sustenance for most of our Fishes, Crustaceans, and other ' sea 
food ', help form Food Yeast, and give us such useful products as 
limestones and the diatomaceous earth {see Chapter II) which is 
widely employed in toothpastes, abrasives, filters, and so forth. 
On the negative side, Algae may be a nuisance in clogging and scum- 
ming and even poisoning fresh waters. 

Among miscellaneous items introduced largely by higher plants 
are the various animal litters and ' farmyard ' manures that form 
such an important part of the farmer's microcosm ; there are also 
the ' green ' manures used, for example, in crop rotation, and the 
organic mulches, etc., that are employed to conserve soil moisture 
and improve soil texture. 

Yet another important product of higher plants is cork, obtained 
principally from the Cork Oak, a tree native to the Mediterranean 
region. Commercial cork consists of the outer bark of the tree and 
can be removed every few years without injury to the tree as it 
grows. Being exceedingly light, compressible but resilient, a low 
conductor of heat and sound, and above all resistant to the passage 
of moisture, it has numerous uses in industry, either in its natural 
form or after being moulded as ' composition ' cork. Among the 
more familiar of these uses are : as stoppers, corkboard, tips of 
cigarettes, and handles of various kinds, in mooring-buoys, lifebelts 
and life-jackets, footwear, tropical helmets, and various sporting 
equipment, and in linoleum and linotiles. 

It should also be recalled that, as implied earlier, green plants 
enable us to breathe by returning oxygen to the air during photo- 
synthesis. They are also the primary source of most vitamins, 
without which we would expire from a combination of deficiency 
diseases. And all the time these same plants afford valuable research 
materials on which many of the greatest scientific discoveries have 
been made and important studies continue practically throughout the 

Some Nuisances 

The significance of various lower plants, particularly, in causing 
diseases of animals and other plants has already been referred to. 
Thus, such human diseases as tuberculosis, brucellosis, tetanus, 
typhoid and some other fevers, plague, cholera, diphtheria and many 
more, are all due to Bacteria, while the Potato and Chestnut Blights 


and Cereal Rusts and Smuts are due to Fungi, which alone cause 
many hundreds of different plant diseases, often of devastating effect 
and vast importance. Other plant maladies may spoil amenities, an 
example being the Dutch Elm disease which threatens the various 
Elms that form such an important feature of the landscapes in 
temperate regions on both sides of the North Atlantic. Some higher 
plants and Fungi, particularly, are dangerously poisonous ; but 
although large doses may be lethal, small ones are often beneficial 
(as we saw when discussing drugs). Consequently such plants are 
not unmitigated curses ; and indeed, provided they are known and 
understood, they may be of real benefit. On the other hand, some 
of the selective herbicides may, like disease-provoking organisms, 
yet be turned to destructive use in biological warfare. 

The saprophytic Fungi and Bacteria that cause harmful decay, 
putrefaction, and often loss of food and fabrics, etc., have also been 
mentioned above ; in many circumstances there would be no organic 
breakdown without them. At all events the beneficial ' scavenging ' 
which these organisms accomplish is so essential as immeasurably 
to outweight such nuisances as they perpetrate, considerable though 
the latter may be — especially with the deterioration and spoilage 
that are so rapid and marked in the tropics. 

Weeds are often defined as ' plants growing where they are not 
wanted ' — which is a broader conception than we had in dealing 
with them in the last chapter. However, the same plant that is 
useful in one place may be obnoxious in another, in which its 
necessary control or eradication becomes a laborious and costlv 
procedure ; in short, it has become a weed. Modern methods of 
weed control include chemical spraying with such herbicides as 
2,4-D or chlorates, mulching, and the biological use of ' smother 
crops ' which combat the nuisances by competition. Other methods 
include the introduction of diseases of the pests involved, as well 
as such time-honoured activities as weeding, hoeing, harrowing, and 
the like. Also effective is prevention of the growth of weeds by 
the use of ' clean ' seed, sterilization of the bed by heat or chemicals, 
and removal of nearby sources of infection. 

Besides being nuisances in the ways mentioned in the last chapter, 
some weeds, such as the Ragweeds and many Grasses, have airborne 
pollen to which people suffering from hay-fever are particularlv 
sensitive. Such plants as Poison-ivy are also a great nuisance to 
some individuals. And finally, not only field and garden crops but 
also forests and even waters have their weeds — in forests the often 


valueless but fast-growing and strangling Birches and Cottonwoods, 
and, in water, the Algae which foul ships' bottoms and reservoirs, 
clog irrigation ditches and navigation channels, and are troublesome 
in many other ways. 

Further Consideration 

If it is felt that the resume of economic botany given in this chapter 
is scarcely appropriate to a work on plant geography, however wide and 
introductory this may be, it should be recalled that the essence of our 
subject is distribution, and that not only the origin and supply but also 
the availability where it is needed of a plant product (and hence its geo- 
graphy) is of importance to Man and pertinent to our study. For Man 
often ' shapes ' plants and their distribution even as they, in turn, largely 
qualify his life and so limit the whereabouts and size of his populations. 

The importance of plants to mankind is stressed in the following works : 

F. O. Bower. Plants and Man (Macmillan, London, pp. xii -f- 365 

R. Good. Plants and Human Economics (Cambridge University Press, 
Cambridge, Eng., pp. xii -f 202 and additional maps, 1933). 

W. W. Robbins & F. Ramaley. Plants Useful to Man, second edition 
(Blakiston, Philadelphia, pp. ix + 422, 1937). 

C. J. Hylander & O. B. Stanley. Plants and Man (Blakiston, Phila- 
delphia, pp. x + 518, 1941). 

J. Hutchinson & R. Melville. The Story of Plants and their Uses to 
Man (Gawthorn, London, pp. xv -j- 334, 1948). 

For further information about most economically important plants : 

A. F. Hill. Economic Botany, second edition (McGraw-Hill, New York 

etc., pp. xii + 560, 1952). 
R. W. Schery. Plants for Man (Prentice-Hall, New York, pp. viii + 564, 


E. E. Stanford. Economic Plants (Appleton-Century-Crofts, New York, 
pp. xxiii + 571, 1934). 

K. H. W. Klages. Ecological Crop Geography (Macmillan, New York, 
pp. xviii + 615, 1942). 

R. Zon & W. N. Sparhawk. Forest Resources of the World (McGraw- 
Hill, New York & London, 2 vols., pp. xiv + 1-493 an< ^ v ^ ~f~ 495~ 

997> 1923)- 
W. W. Robbins. The Botany of Crop Plants, third edition (Blakiston, 
Philadelphia, pp. x -f 639, 193 1). 


United States Department of Agriculture 1950-51 Yearbook. Crops in 
Peace and War (U.S. Government Printing Office, Washington, 
D.C., pp. [xviii -j-] 942, 195 1). A veritable mine of information 
inter alia on chemurgical uses and future possibilities. 

Poisonous plants are commonly treated on a local basis : 

L. H. Pammel. A Manual of Poisonous Plants : chiefly of eastern North 

America. . . (Torch Press, Cedar Rapids, Iowa, Pt. I, pp. viii + 150, 

and Pt. II, pp. 153-9:7, I9 11 )- 
J. \V. Harshberger. Textbook of Pastoral and Agricultural Botany 

(Blakiston, Philadelphia, pp. xiii -f 294, 1920). Gives some idea of 

the magnitude of the subject on a wider basis. 

As well as the above more general works, there are various books which 
are devoted to particular crops or closely-allied groups of crops. Along 
these lines, apart from individual works, there are three special series in 
English of which (1) the Economic Crops series of Interscience Publishers, 
New York, has so far published books on Apples, Bananas, Cherries, 
Cocoa, and Sweet Corn, while in (2) Longmans Green & Company's 
Tropical Agriculture series there have appeared books on Cocoa, Rice, 
Tea, and Bananas ; in preparation are Coconuts, Rubber, Oil Seeds, 
Sorghum, Oil Palms, Cotton, and Spices. Also extensive and uniform 
will be (3) the World Crops Books series of Leonard Hill Limited, 
London, in which the volumes so far in press or arranged include separate 
ones on Alfalfa, Barley, Brassicas, Coffee, Cucurbits, Eucalyptus, Flax 
and Linseed, Hops, Jute, Mangoes, Mushrooms and Truffles, Oats, 
Onions and their allies, Peanuts, Pineapple, Rubber, Rye, Sugar Cane, 
Taros and their allies, Tomatoes, Tropical Cash Crops, Vegetable Fibres, 
and Wheat. Each of these World Crops Books deals with the botany, 
cultivation, and utilization throughout the world of the crop or group of 
crops concerned, and each is profusely illustrated and fully documented. 
An allied series, entitled Plant Science Monographs, also published by 
Leonard Hill Limited, deals especially with research advances involving 
inter alia many of these and other important crops. 

Chapter X 

We have now dealt sufficiently with special kinds and systematic 
groups of plants and their distributions, and must proceed to 
consider the results of their natural aggregation, namely, vegetation. 
As a basis for this we must deal in the present chapter with ecological 
factors and, in the next chapter, with the main habitats these factors 
constitute. In addition there will be considered in Chapter XI 
certain fundamental tendencies and attributes of vegetation whose 
recognition is essential to the full understanding of vegetational 
changes and types. 

Ecology is the study of the mutual relations among organisms and 
between them and their environment — environment being the aggre- 
gate of all external conditions and influences affecting life and 
development at a given spot. Ecological or environmental factors 
are many and diverse, and often intricately mixed and interdependent. 
Either singly or in combination, the various ecological factors may 
affect the presence or absence, vigour or weakness, and relative 
success or failure of various plant communities through their 
component taxa. Although the subject is enormously complex, the 
immediate vehicles of influence are very few, being chiefly food, 
light, temperature, water, and dissolved substances ; these are 
affected by variations in the ecological factors which characterize 
different habitats and consequently lead to the differentiation of 
vegetational types. 

The four main classes of ecological factors with which we shall 
deal below — namely, climatic, physiographic (of topography, etc.), 
edaphic (of soil), and biotic (due to living organisms) — are them- 
selves commonly interrelated and intricately mixed. Often they 
work through one another, acting and reacting together, as in the 
case of physiographic changes which bring about local climates that 
in turn may affect the soils and competition-impress. Accordingly 
this classification, like so many other biological ones, is to some 
extent artificial. Yet all categories affect the plant either directly 
or indirectly through modification of its reactions, bringing about 



functional responses or differences in growth or structure, although 
different plants vary widely, of course, in the nature and degree of 
such response. 


The climatic factors comprise the general features of regiona 
climate, often being rhythmic — exhibiting, for example, diurnal, 
seasonal, or long-term cyclic fluctuations. They may also vary 
locally, to give local climates, and even do so in extremely restricted 
confines, to give microclimates. Examples of local climates are 
found on steep northern or southern hill-slopes, and of microclimates 
on the leeside of boulders which protect the immediately neighbour- 
ing plants and animals from wind and insolation. In general the 
factors classed as climatic have a dominating influence. Neverthe- 
less, so far as plants are concerned, these factors are often but 
poorly expressed by meteorological records which, for example, are 
usually taken at some standard height above the ground (rarely the 
height of the plant) and fail to observe the often rapid fluctuations 
which can be so important to a sensitive organism. There are five 
main climatic factors which we must now consider in turn: it is 
their different combinations which largely prescribe vegetation-types, 
and so they cannot well be placed in order of relative importance. 

(1) Light. This, as we have seen, is essential for photosynthesis, 
though fortunately sufficient illumination for this purpose is present 
almost everywhere on land and in surface waters ; it may also be 
important for some reproductive processes. Fig. 79 shows the 
Moss Campion (Silene acaulis agg.) flowering only on the top and 
south-facing side of its domed tussocks in high-arctic Spitsbergen, 
where it can constitute a reliable compass ; the effect is thought to 
result from differing light intensities. 

The light-climate at a spot depends on the duration, time- 
distribution, intensity, and quality of the insolation there obtaining, 
though so far as the plant is concerned the effective period may be 
modified by cold or drought. Moreover, for successful flowering 
many plants require a relatively long (or, alternatively, short) day, 
apparently regardless of the intensity or quality of the light, and 
consequentlv such plants are largely limited to high (or low) latitudes, 
respectively. The effect of light on photosynthesis depends largely 
on intensity, which also influences growth. In the open, elongation 
is checked and lateral organs enlarge, whereas with congested 




conditions, for example in forests, the form tends to be more elongated 
and narrow. In temperate forests, too, the seasonal aspects are 
apt to be important — in particular the prevernal {i.e. before spring) 
one of herbs which flower before the shading tree-leaves expand. 
Thereafter different levels or layers in the forest usually have 
different light-climates. 

The measurement of light tends to be unsatisfactory, for no instru- 
ment indicates precisely the quality as well as intensity, much less 
the total quantity ; usually only the intensity at a particular point 

Fig. 79. — Moss Campion {Silene acaulis agg.) flowering only on the south-facing 

sides and tops of its domed tussocks near 8o° N. in Spitsbergen. The surrounding 

terrain is a mixed ' half-barren '. 

in time and space is measured, and that is done only as far as the 
measuring device employed is sensitive to the component wave- 
lengths encountered. The particular rays of the spectrum which 
are most effective in the diverse plant-functions affected by light, 
also tend to be different. Nevertheless photoelectric cells and 
actinometers of sensitized paper are useful, especially for purposes 
of comparison, and both are widely employed by ecologists in the 

(2) Temperature. This factor is vitally important as it conditions 
the speed of the chemical actions and activities comprising life. 
The great world vegetational zones, like altitudinal ones, depend 
primarily on temperature, and we find it convenient to distinguish 
between megatherms (plants favouring warm habitats), microtherms 
(plants favouring cold habitats), and the intermediate mesotherms. 



Different plants are variously adapted as to the minimum, optimum, 
and maximum temperatures for their life as a whole as well as for 
its component physiological functions, even though these actual 
temperatures may change with variations in other conditions and 
with the state of the plant (as well as, of course, differently with 
different plants). 

Winter is normally a resting period when activity is at a minimum 
in temperate regions, though many plants are active at much lower 
temperatures in the polar lands and waters — some even below o c C. 
On the other hand, temperatures above the freezing point may 
already be lethal to tropical plants ; so may temperatures above 
45 C. if evaporation does not cool and save them. However, there 
can be few places that are naturally too hot or too cold for any 
plants. Important indirectly are clouds and other influences reduc- 
ing the amount of direct insolation, and relative humidity w T hich 
greatly affects the loss of water by evaporation. The soil also has 
a marked local effect, dry and dark types warming up much more 
quickly than heavy waterlogged ones. 

Temperatures vary markedly at different levels as well as at 
different times, and meteorological means (i.e. averages) are there- 
fore of little value to the ecologist. An annual mean well below 
freezing point is found in some continental regions where forests 
abound, and although monthly means, and especially monthly mean 
maxima and minima, are of more value to the plant scientist than 
annual means, a thermograph tracing showing the continuous change 
on the spot is most desirable for ecological purposes. To give any- 
thing like a complete picture of the temperature-climate as it strikes 
the plants, tracings should be obtained synchronously at each 
different level or layer of vegetation and root-infested soil. In 
intricate work involving, for example, the surfaces or the internal 
tissues of leaves, thermocouples are employed instead of ther- 
mometers, while for determining the approximate temperatures of 
hard surfaces, etc., thin shavings of paraffins of different known 
melting points are convenient. Fig. 80 indicates (A) the annual 
mean temperature, and (B) the mean temperature of the warmest 
month of the year, in different parts of the world. 

(3) Precipitation. The amount of rain, especially, falling in an 
area during the year constitutes a factor of outstanding importance, 
as it often mainly determines the availability of water for growth 
and other vital processes. To this availability the local vegetation 
largely corresponds ; and although the year's total is apt to be the 


feature most important for trees, the season in which it falls may 
matter a great deal to herbaceous plants and grasslands. These 
last are especially favoured by spring rainfall in regions of cold 
winters. With hot and dry summers but winters warm enough 
for growth, there may be a preponderance of sclerophyllous shrubs 
{i.e. having rather small leathery leaves). With less and less rainfall 
there tend to be still more xeromorphic plants {i.e. with features 
aiding them to conserve water, as illustrated in Fig. 20, xerophytes 
being in general plants which grow in dry places). At the other 
extreme are hydrophytes, living more or less submerged in water 
(Fig. 21), and hygrophytes, or marsh plants, while between xerophytes 
and hydrophytes lie the so-called mesophytes, living in habitats that 
usually show neither an excess nor a deficiency of water. 

Rain is caused by the cooling of moisture-laden air. Rainfall is 
usually reported in the form of monthly means, which are the 
amounts falling in the various calendar months but averaged over a 
period of years, though the number of rainy days in each month 
gives a better indication of its distribution. Moreover, sudden 
heavy rain is apt to be largely lost as run-off, and may cause bad 
erosion. Because of the often marked local differences with physio- 
graphic changes, it is desirable for an ecologist to have his own 
automatic rain-gauge which, like his thermograph, needs tending 
only once a week. 

Other forms of precipitation are snow (which may lie on the 
ground to form a valuable protective blanket and also a reservoir 
of water, but is apt to limit the growing-season by its late melting), 
hail (which may cause serious injury especially to young crops), 
sleet, and dew (which is important in some deserts where it pro- 
vides much of the surface water on which the ephemeral plants 
depend). Fig. 81 indicates the average annual precipitation in 
different parts of the world. 

(4) Evaporating power. The evaporating power of the air is a 
factor of the utmost importance to the life of plants, as it directly 
affects their transpiration. It is indicated approximately by the 
relative humidity (the ratio of the water-vapour present in the 
atmosphere to that necessary for saturation at a particular tempera- 
ture), or, more accurately, by the ' saturation- deficit ', which takes 
account also of temperature, and determines the ' pull ' exerted by 
the atmosphere on the water-economy of plants. 

The relative humidity is commonly measured by means of a 
' wet and dry bulb ' hygrometer (psychrometer), the difference in 



temperature of the wet and dry thermometer bulbs giving a measure 
of the deficiency of water-vapour below saturation point in the air 
tested, while the saturation-deficit can be calculated directly from 
these different readings. Although this instrument is customarily 
swung through the air to simulate wind and replace any saturated 
layers that may form around the wet bulb, from our point of view 
it is best kept stationary as plants are. Even the hygrometer gives 
readings only at a particular time ; often it is more important to 
know the effect over a given period, and this may be measured by 
using an ' atmometer ' (evaporimeter), which indicates the water- 
loss bv evaporation from a given area of porous pot. This integrates 
the water-vapour content of the air with temperature, wind, and the 
time-factor, and may be compared with a plant which it is placed 
beside. Or, for continuous recording, a tracing may be made by 
an automatic hygrograph, such as those using strands of human 
hair which is highly sensitive to changes in atmospheric humidity. 
Batteries of such instruments will often show marked differences in 
different strata of a forest, for example — indicating local variation 
which may be of great importance in varying the ' pull ' exerted on 
different plants, on the same plant at different times, or even on 
different parts of the same plant at a particular time. 

(5) Wind. Owing to the friction of the soil surface, rocks, 
buildings, and above all major physiographic features and masses 
of vegetation, winds tend to increase in velocity with height above 
the ground. Wind commonly affects other ecological factors in a 
given spot — for example, water content and temperature, through 
its effect on evaporation — but can also have a direct influence on 
vegetation, especially by uprooting trees or by breaking off branches 
or other portions. It has a similar, usually drying, effect upon the 
soil, or may occasionally act in the opposite direction by bringing 
up moister air which reduces transpiration and evaporation, and 
may actually lead to deposition or precipitation. Most widely 
important to plants, however, is the manner in which wind increases 
water-loss, by constantly bringing unsaturated air into contact with 
leaves and young shoots. Mechanically, wind can also cause 
erosion of soil and abrasion of vegetation through carriage of 
particles, and physiologically it can decrease growth by replacing 
damp by dry air, and consequently increasing the transpiration and 
reducing the turgor of organs on which it impinges. This explains 
the frequent growth of trees and shrubs chiefly away from the 
direction of the prevailing wind in exposed situations (cf. Fig. 82, A), 


and their restriction to tangled dwarfs in sheltered depressions 
(Fig. 82, B). In strong dry winds young parts of plants may even 
become shrivelled and killed in a few hours, and surface soil may 
become dried out. Such effects may be observed in the warm foehn 

Fig. 82. — Effect of wind on trees. A, deformed outliers of Black Spruce (Picea 
man ana) near Churchill, Hudson Bay. The prevailing winds are from the left. 
Note in foreground the luxuriant prairie-like tundra of the arctic-subarctic ecotone. 
The lower branches of such Spruces tend to form a dense layered mat that is well 
protected from desiccation in winter by drifted snow, the farthest outliers (on 
left) often being limited to such growths. Upper branches (on right) ' trail 
away ' to leeward. B, deciduous broad-leafed and evergreen coniferous tree 
species forming a dwarfed tangle in sheltered depressions on exposed hill-side on 
coast near North Berwick, Scotland. The Grasses and other herbs reach the 
height of the gnarled ' trees ' which scarcely anywhere exceed the level of the sur- 
rounding terrain. 


or chinook winds that, sweeping down from mountains, can raise 
the temperature of the air locally by as much as 30 C. in a very 
short time. For such reasons a good deal can be inferred about 
the wind-climate of a habitat by direct observation of the vegetation, 
etc. Wind velocity is measured by an anemometer, but its effect 
is included in observations obtained from stationary hygrometers 
and atmometers. 

As we saw in Chapter IV, wind is an important agent of dispersal. 
It may also be of significance phytogeographically in determining 
the local distribution of species or communities of plants, some 
types being markedly more wind-resistant than others, and many 
being unable to flourish or even exist in exposed situations. Thus 
the tree-limit on the sides of mountains is apt to be due largely to 
winds, as may be seen by the frequent persistence higher up of 
groups of trees in sheltered pockets ; such trees further reduce 
exposure verv locally and enable tender herbs to grow in their 

Ocean currents, such as the warm Gulf Stream or the cold East 
Greenland Current, may have a considerable effect on temperature 
both locally and on land at a distance — especially when the winds 
are predominantly on-shore. By bringing in fresh materials, such 
currents may also alter the conditions with regard to nutrient salts, 
etc. Moreover, as we saw in Chapter IV, water currents can be 
an important aid in plant dispersal. 


The physiographic factors are those introduced by the structure, 
conformity, and behaviour of the earth's surface — e.g. by topographic 
features such as elevation and slope, by geodynamic processes such 
as silting and erosion, and consequently by local geology. Other 
causes of physiographic change from place to place include the 
blowing of sand or dust which in time or special circumstances can 
assume vast proportions. It is in such connections that the various 
landforms described in Chapter XVII tend to be most significant 
to students of plant geography and ecology. 

Physiographic factors act on local vegetation largely through 
climatic or edaphic features which they engender. They are 
consequently sometimes classed with these other groups. Yet, 
ecological factors being so widely interdependent in any case, it 
seems more conducive to clear understanding to consider the 


physiographic ones separately — especially as they are well marked 
in their effect on vegetation in regions of drastic topography and 
harsh climate. Strong topographical relief tends to produce marked 
local climates, summits for example being very different in these 
respects from sides of mountains, and narrow valleys from open 
plains. Quite apart from the tendency to greater windiness and 
exposure at higher altitudes, the air and soil temperatures tend to 
get lower and the relative humidity greater as we ascend, with 
atmospheric pressure decreasing and heat-radiation increasing in 
intensity. Altogether, climatic variation becomes more and more 
extreme and rapid with increasing altitude. 

As an example of physiographic effects in arid regions, we may 
rise from unproductive plains to fertile slopes and forests on moun- 
tain sides, and at very high altitudes reach again an unproductive 
zone of low absolute humidity and rigorous exposure. The changes 
are due largely to local climate but would not take place if it were 
not for the physiography. Moreover, major topographic features 
often affect the climate at a considerable distance — as, for example, 
mountain ranges which may cause rainfall locally and decrease it 
in their lee — while in Chapter IV we saw how such features, and 
wide expanses of water, can act as barriers to plant migration. 

Another important physiographic effect is aspect : in the northern 
hemisphere, north-facing slopes tend to be more hygrophilous 
(adjusted to moist conditions) than south-facing ones at similar 
altitudes (Fig. 83, A). This is owing to the effect of insolation on 
air and soil temperatures, and consequently on relative humiditv 
and evaporation and, through them, on the local water situation 
(even when precipitation is the same). For vegetational differences 
due to topography are most often correlated with moisture, and 
naturally tend to be marked chiefly where water is deficient and 
consequently is a critical factor. Sometimes for this reason there 
may be entirely different vegetations and even floras on the two 
sides of a deep valley or steep mountain (Fig. 83, B), as in the dry 
Mediterranean region, while if similar zones are developed they 
tend to be at higher altitudes on the south- than on the north-facing 
slope. The western slopes of mountains may be noticeably warmer 
and drier than the eastern ones, owing to the sun's afternoon warmth. 
Quite drastic effects may often be seen already on a microclimatic 
scale, for example in the shelter of rocks or even stones on exposed 
sea-cliffs and mountain summits. Here we mav refer again to 
Fig. 79, which shows a striking aspect effect in the Arctic. 

Fig. 83. — Aspect effects in Colorado and Nepal. A, Colorado, U.S.A.: the 
relatively damp .and cool north-facing slope (on left) is covered by a stabilized 
forest of Douglas Fir, whereas the south-facing slope supports mainly dry Oak 
scrub and scattered Ponderosa Pines. (Phot. G. E. Nichols. By permission 
from Plant Ecology, by Weaver & Clements, copyright date 1938, McGraw-Hill 
Book Co.) B, Jumla, West Nepal, c. 3048 metres (10,000 feet) in the Himalayas: 
the north-facing slope (on left) is forested with Abies spectabilis and Betula utilis, 
and, lower down, some Pinus wallichiana, whereas the dry and sunny south- 
facing slope supports chiefly herbs and Grasses, with some scattered low Juniperiis 
bushes. (Phot. O. Polunin.) 



Important in itself is the steepness of a slope, for it largely deter- 
mines the stability of the surface and retention of water, and also 
the effect of aspect or exposure — especially in the higher latitudes. 
Thus in the northern hemisphere a steep south-facing slope will 
receive the strong midday sun's rays more or less perpendicularly, 
while a steep north-facing slope may receive only oblique and weak 
morning and evening rays, or perhaps none at all. These differences 
often have a marked effect, especially on the water and temperature 
conditions in the two places, and consequently on the vegetation. 
For besides the intensity, the duration, quality, and quantity of 
incident light are at the same time affected. 

Slope can also greatly affect the character as well as the amount 
of soil which accumulates. This, like the nature of the underlying 
rock, is often reckoned as an edaphic factor ; but in so far as either 
determines or results in topographic change, it is to be considered 
as physiographic. Different textures and types of rocks will produce 
different topographies which bring about local climatic differences 
that are physiographically engendered. Such differences may also 
affect the water conditions, including the level below which the 
ground is waterlogged or frozen, and so again drastically affect 
the habitat. 

Geodynamic agencies are particularly active in mountain districts 
and about coasts, causing all manner of changes in topography — 
sometimes almost from day to day. Steep slopes and river banks 
are constantly being eroded, the material sliding down as talus, etc., 
or being washed down and deposited elsewhere. Such activities 
often cause the formation of new ' open ' habitats, both in the 
places whence the eroded material came and in the areas in which 
it comes to rest. Frosts may help disintegration through their 
heaving, splitting, and other erosive tendencies, while avalanches 
often clear off the surface materials from considerable areas. And 
even as mountain slopes change their surfaces, and river-beds alter 
their outlines and vary courses, so do sea-coasts and cliffs vary in 
their conformation, erosion being widely at work. In other places 
there are silting salt-marshes or gradually moving sand-dunes or 
shingle-banks — to mention only a few of the geodynamic sources of 
physiographic change. 


The edaphic factors are those which are dependent on the soil 
as such — on its constitution, water and air content, inhabiting 


organisms, and so forth. We have seen how climatic factors, such 
as temperature and precipitation, are of supreme importance in 
determining the general character of the vegetation over wide areas ; 
here and in the next section we shall deal with the edaphic and 
biotic factors, respectively, which are apt locally to modify the 
conditions and vegetations of these major climatic belts. Especially 
has it long been recognized, and utilized in agricultural, horticultural, 
and forestral practice, that differences in the soil are often largely 
responsible for differences in vegetation within the same climatic 
region : consequently they are of great significance in plant 

Soil may be considered as the unconsolidated superficial material 
of the earth's crust, lying below any aerial vegetation and undecom- 
posed litter, and extending down to the limits to which it affects 
the plants growing about its surface. Beneath the soil lies the 
subsoil or unaltered rock. Though usually composed primarily of 
material derived from the parent rock, the soil has come into being 
largely through interaction of this ' substratum ' with climate and 
living organisms. Thus its texture may be dependent largely on 
water- and frost-action and other ' weathering ' tendencies, while its 
content of humus (partially decomposed organic matter) results from 
the contributions and activities of inhabiting plants and animals. 
For soil is a veritable ' microcosm ' or little world — with its own 
physical structure, chemical composition, atmosphere, flora, and 
fauna. And characteristically it exhibits a perpetual series of actions 
and reactions between organisms and environment. In it about 
one-third (by volume) of the bodies of higher plants spend their 
lives — influencing, and being influenced by, the particular conditions 
obtaining in the soil. These conditions, through the subterranean 
organs of plants, in turn influence their living aerial parts. 

Soils that are undisturbed by agriculture and other factors com- 
monly become stratified into layers or ' horizons ' at different 
depths, which often have very different compositions as well as 
natures. Such ' profiles ' are, however, largely destroyed by cultiva- 
tion. Normally, three main horizons or groups of horizons are 
exhibited, namely, the upper or ' A ' zone of extraction of soluble 
salts and fine-grained materials, the middle or ' B ' zone of their 
concentration, and, below, the ' C ' zone where neither extraction 
nor accumulation has occurred at all extensively. The characteristic 
profiles so formed are to a considerable extent climatically engendered, 
and so mature soils can be classified broadly into climatic types or 


' world groups ', such as podzols (developed- chiefly in cool regions 
of high precipitation relatively to evaporation), brown earths (with 
generally lower rainfall and higher temperatures), chernozems (with 
low rainfall in continental regions), prairie soils (with higher rainfall), 
chestnut -brown soils (in warmer and drier places), laterites (with high 
rainfall in the tropics), red loams (with lower rainfall in warm- 
temperate regions), tundra soils (of polar regions where the subsoil 
remains frozen and organic decomposition is retarded), and so on. 
Fig. 84 shows three characteristic soil profiles and Fig. 85 indicates 
the distribution of the primary soil groups of the world. 

As we shall see towards the end of the next chapter when dealing 
with plant succession, soil development and vegetational develop- 
ment are intimately connected, both being largely controlled by 
climate. Meanwhile the essential constituents of most soils may 
conveniently be treated in five categories : 

(1) Mineral fragments of various sizes resulting from the dis- 
integration of rocky materials by physical and chemical weathering ; 
the parent material may be either local or otherwise, as may be the 
weathering. These mineral constituents form the inorganic frame- 
work and differ widely according to the physical and chemical 
nature of the parent material. They affect plants particularly by 
bringing about variations in soil water and aeration, as water reten- 
tion is much affected by mechanical composition (the relative 
proportions of different-sized mineral particles present). The 
mineral constituents may also affect the composition of the soil 
water in important ways {see below). 

Soils are mechanically analyzed by separating into ' fractions ' the 
particles whose sizes lie within definite arbitrary limits, ranging from 
larger stones and gravel (more than 2 mm. in diameter) down 
through various sizes of sand and silt to clay (less than 002 mm. 
in diameter). The majority of mature soils consist largely of silica 
and silicates which are relatively insoluble and form a more or less 
permanent basis, any calcareous material tending to become dis- 
solved (' leached ') out of the surface layers and many of the finer 
insoluble particles being carried down mechanically to lower levels 
(this process is termed ' eluviation '). Particularly important in 
leaching and chemical weathering is the acid-forming carbon dioxide 
dissolved in soil water. The particles forming the soil's inorganic 
framework tend to be coated with colloidal material of very fine 
clay or of organic origin, which may help cement them into com- 
pound particles or grains and increase the ' crumb structure ' of the 




soil and its ability to hold water and salts. Sandy soils are porous, 
' light ' to work, easily penetrated by roots, and they dry readily ; 
in contrast, clayey soils are retentive of water, heavy, poorly aerated 
and sticky when wet, and hard when dry. For most working and 
plant-growing purposes, mixed ' loams ' are best. 

(2) Soil water containing dissolved substances is also of funda- 
mental importance, being commonly the chief source of water for 
plants. Water is of course essential to plants as usually their main 
constituent by weight, as the medium for physical and chemical 
changes, and because large quantities must be absorbed to cover 
the continual loss by transpiration from their surfaces. Except in 
extremely dry soils and below the level of permanent ground-water, 
the soil water mainly forms films around the component particles 
of the soil. The amount of such water and thickness of the film 
depends on such factors as the soil's mechanical constitution, on the 
recency of precipitation and tendency to run-off, on subsequent 
weather conditions, on humus content, on the covering of vegetation 
and litter, etc., and on the effect of this covering on water loss 
by transpiration and evaporation. Especially does water percolate 
through and evaporate from coarse gravelly or sandy soils, and remain 
suspended in fine clayey and humous ones with their high capillary 
action and immense aggregate surface of microscopic or colloidal 
particles. However, in such retentive soils much of the water may 
be so strongly held that it cannot be abstracted and used by the plants, 
so that it is often necessary to distinguish between water which is 
available to plants (the so-called chresard, above the point of per- 
manent wilting, though this may vary somewhat with different 
species) and that which is so strongly held as to be unavailable 
(echard). The entire water content of the soil may conveniently 
be termed the ' holard '. Water tends to promote the stratification 
of soils, and the soil's content of available water is often the chief 
factor causing local differences between plant communities. 

Another important factor affecting vegetation is the soil water's 
content of dissolved inorganic salts, etc., which are derived from 
the mineral matter present and from organic breakdown. Certain 
of these salts' constituent elements are essential to the plants' con- 
tinued well-being and even to their life, while others are apt to be 
obnoxious or actually poisonous. Outstanding are the extremely 
saline soils which are inhabited only by specially adapted plants 
(halophytes). In these connections the various needs and abilities 
of different plants lead to correlations with the chemical content of 


the soil water, which frequently constitutes' a determining factor in 
plant geography. Examples occur in the cases of plants which seem 
to require ' lime ' (calcicoles or calciphytes) and those which appear to 
avoid it (calcifuges or oxylophytes), though frequently such prefer- 
ences are bound up rather with questions of basicity or acidity, 
respectively. For the soil ' reaction ' (hydrogen-ion concentration) 
also can determine the presence or absence of particular species 

Thus can the varying tolerances of many plant species, etc., to 
different environmental factors come into play as particularly 
important phytogeographically, though it should be remembered 
that away from their optimum sphere plants tend to be more and 
more susceptible to competition and other influences, and hence to 
show less and less ability to persist. Consequently it is especially 
towards the limits of their ranges that they are liable to be most 
narrowly restricted to special habitats involving particular environ- 
mental conditions. 

(3) Soil atmosphere mainly occupies the interstices between the 
soil particles or crumbs, with their covering films of water. It 
tends to contain a slightly lower proportion of oxygen and a much 
higher one of carbon dioxide than ordinary air, and to be normally 
saturated with water-vapour. This may not be the case in the 
surface layers of very dry soils, while at the other extreme water- 
logged ones are liable to be deficient in oxygen, their lack of aeration 
making them unsuitable for most forms of plant and animal life. 
Normally, plentiful oxygen in the soil is necessary for the life of 
most of the microorganisms and other inhabitants and for the 
respiration of the underground parts of higher plants growing in 
it, the differences in composition from the free air being reflective 
of the gaseous exchanges involved, namely, absorption of oxvgen 
and giving out of carbon dioxide. However, interchange with 
the free air by diffusion and other agencies appears to be fairlv 

(4) Organic matter, arising from the death of plants or parts of 
plants, or of animals or added as manure, forms another highly 
important constituent of most soils. Included are roots that decom- 
pose in situ. Indeed all soils which bear vegetation (and according 
to some authorities all true soils) contain dead organic matter, 
usually more or less broken down to humus, though the amount 
may vary from very little in fresh ' new ' soils to virtually 100 per 
cent, of dry weight in peat and leaf-mould. In the dissemination 


and breakdown of the dead plant and other remains, Earthworms 
may be important in dragging down and partly digesting the material. 
After this, soil Fungi and Bacteria cause further disintegration and 
ultimately decomposition into more or less simple salts (such as 
ones containing nitrogen and phosphorus which are essential to 
plants), carbon dioxide, and water. These are the fundamental 
plant foods, and the organic matter is the chief seat of the activities 
of the microorganisms liberating (and sometimes producing) them. 
Such decomposition and disappearance takes place relatively rapidly 
in warm, moist, and well-aerated soils, these conditions being 
favourable to the activity of the ' scavenging ' organisms. On the 
other hand, in cold and wet soils that are poor in salts and acidic 
in reaction, the debris may long remain on the surface as scarcely 
decomposed ' raw humus ' (mor). 

Ordinary humus, having become structurally unrecognizable and 
largely colloidal, improves both heavy clay and light sandy soils, 
lightening the former and giving consistency and water-holding 
capacity to the latter, as well as adding plant food in each case. The 
' reaction ' of a soil — its degree of acidity or basicity — is also import- 
ant to many plants and may affect their distribution, as particular 
species are apt to show a distinct preference for soils whose reaction 
lies within certain limits. This reaction again is largely bound up 
with humus content, humus being the main source of acidity in 

(5) Living organisms, together with the roots and other under- 
ground parts of aerial plants, comprise the other widely essential 
constituent of the soil. They include, as we have already seen, 
the mainly microscopic soil flora and fauna whose life is usually 
centred on the soil's humus content and which are often very 
sensitive to changes in the conditions of their limited environment. 
They also include the main mixing and ' scavenging ' animals and 
above all saprophytic Fungi and Bacteria, etc., which are so vitally 
important in the maintenance of ecological balance and indeed of 
life in the world. For higher plants, as we have already noted, 
these organisms make available various essential food substances, 
including nitrogen in usable form. 

In highly acid soils, Fungi largely replace the decomposing and 
other useful Bacteria — such replacement in itself involving significant 
phytogeographical changes. Indeed these soil communities and 
their component organisms have their distributions and other 
geographical implications in much the same manner as higher types. 


Besides various essential or beneficial substances including growth- 
stimulating ones, harmful toxins may be produced in the soil by 
living organisms or parts of organisms — with corresponding effects 
on plant distribution. And then there are various mycorrhizal 
associations, Algae, and other units in the microcosm, all of which 
may have their importance to the vegetation and to plant geography, 
though quite how is often not well understood. Nor should we 
forget the importance of soil temperature, which can act in so many 
and various ways — as, for example, through the soil's living content. 


The biotic factors in the wide sense are those due to living 
organisms, whether animal or plant — ranging, as it were, from Man 
and the great herbivores and trees down to the lowly but often 
vitally important soil microorganisms with which we have just dealt. 
It is useful to visualize the total components of an immediate 
environment or recognizable habitat as forming a self-contained 
ecosystem, composed on one hand of the inorganic and dead parts 
and, on the other, of the various organisms which live together in 
it as a sociological unit and comprise the biota. A large, primary 
biotic community in which the climax vegetation {see next chapter) 
is more or less uniform is termed a biome. Our interest is primarily 
in the living components, the inert ones being from our point of 
view mere factors conditioning the existence, structure, and develop- 
ment of the biome — chiefly through their effect on one or more of 
its component organisms. These are the individual species or other 
taxa, divisible approximately into animals and plants. The latter 
form the plant community, which may be loosely defined as an 
entire population of plants growing together and maintaining as a 
whole a corporate individuality that is not the same as the sum 
total of the separate manifestations and effects of its components. 

The component plants of a community have many immediate 
internal (' autogenic ') effects upon one another and upon their own 
habitat, as for example in competition and the deposition of humus, 
and by producing various changes in the soil ; they even have 
1 allogenic ' effects in sheltering other plants and in dispersing them- 
selves outside the immediate community. It is convenient, however, 
to treat these plant-engendered repercussions under succession (see 
next chapter) and to regard separately as introducing the (collective) 
biotic factor chiefly those animals which have a marked effect upon 


the community or any part thereof. These are largely external in 
origin. For practical purposes such factors manifest themselves in 
the totality of direct and indirect effects of animals on plants. It 
is, however, also convenient to include with the animals such plants 
as the lowly scavengers and agents of chemical change, and, in 
addition, those which are damagingly parasitic. Also mentionable 
here are ' insectivorous ' (carnivorous) plants, climbers, epiphytes, 
and mycorrhizas — which last, like Lichens and other symbioses, are 
concerned with the existence of more than one species at a point. 

Apart from changes engendered by the main or subsidiary com- 
ponent plants themselves, organisms of many other sorts may affect 
a plant community in numerous and diverse ways. There are the 
soil Bacteria, causing all manner of important chemical changes ; 
the Protozoa, which devour the Bacteria ; the Earthworms which 
help disintegrate organic matter and aerate the soil ; the Fungi 
which carry this breakdown further ; the Snails which eat plants ; 
and the browsing Mammals which may have the most profound 
effect — for example in turning forest into treeless pasture. Other 
important effects may be produced by Man and his domestic animals 
and fires, by the Beavers which fell trees and turn valleys into lakes, 
by the caterpillars and Locusts which may devastate whole areas, by 
the Insects and Birds which pollinate flowers and carry diseases, 
by the various animals which are so important in dispersing seeds 
and other disseminules, and by the parasitic Fungi, etc., which in 
extreme cases may even kill the dominant plant and change the 
whole aspect locally. 

Parasitic lower plants (or sometimes invertebrate animals or 
vascular plants) are, as we have already seen, very important to 
crops whose mode of growth often encourages them or their ' carry- 
ing ' insects. Less known is their effect on ' natural ' vegetation ; 
seemingly it tends to be less drastic as equilibria are approached, 
though we still do not know what organisms may be kept out by 
potential predators, or perhaps so weakened in competition as to 
fail to become established. Certainly the death-rate of seedlings 
tends to be very high. If we had not plentiful records and even 
recollections of the Sweet Chestnut forests of eastern North America, 
how could we know that they had been devastated by blight and 
that their absence nowadays is not due to some inimical climatic 
or other factor ? And is it not possible that some other plants whose 
remains tell us that they once flourished in areas where they do not 
now grow, were ousted by parasites or other biotic impress rather 


than by the climatic vagaries which are indubitably the usual cause 
of such major distributional change or extinction ? The fan-shaped 
and ' feathery ' Elms of New and Old England, respectively, which 
are such an important feature of the landscape of those two delightful 
parts of the world, are threatened by the so-called Dutch Elm disease 
which has already considerably changed the local environment in 
many areas. The White Pine Blister-rust is apt to kill the dominant 
species of some of the most widespread and important forests of 
North America. These are but a very few instances of devastating 
parasitic attacks on natural or semi-natural vegetation. 

The small herbivores include Snails, Slugs, Locusts, and the larval 
stages of Insects such as caterpillars (the larvae of Butterflies and 
Moths). These often cause considerable local damage and even 
devastation — particularly to individual crops but sometimes to 
native species or whole tracts of vegetation. 1 But their exclusion 
even for experimental purposes is often difficult, taxing the ingenuity 
of the investigator. The effects of the smaller vertebrate animals, 
such as Voles and Mice, tend to be most conspicuous where Man 
has upset the balance of nature by destroying their enemies, such 
as Hawks and Foxes. Rabbits may be especially troublesome, often 
converting heaths or even forest into grassland by their gnawing and 
prevention of regeneration of such dominant trees as Beech on chalk 
in England. This is easily demonstrated experimentally by effec- 
tively wiring off areas (the netting must be sunk some inches in 
the ground) which soon come to exhibit the early stages of forest 
regeneration, whereas the pastured surrounding tracts remain close- 
cropped and grassy. Care must be taken also to exclude small 
Rodents and Birds which can destroy tree seedlings, for example, 
by nipping off the tender young shoots. It will be interesting in 
due course to observe the effect on the local vegetation of the virtual 
(if only temporary ?) extermination of Rabbits by myxomatosis in 
several European countries. 

Of large browsing Mammals such as Buffaloes, which used to 
inhabit the great grassy plains, the natural populations are now 
widely destroyed. They were probably important in the mainten- 
ance of the grasslands, at least in the damper areas where trees are 
able to grow, but are nowadays largely replaced by domestic herds, 

1 It has even been suggested that a plague of caterpillars may have been a 
leading cause of the dying out of the ancient Norse settlements in West Greenland 
— by devouring the vegetation and hence starving the Sheep on which the Norse- 
men were largely dependent. 


which have a similar effect upon the vegetation. Thus the forest 
generally tends to advance on the grassland with the removal of the 
heavy pasturing which ordinarily keeps it in check. Commonly, 
browsing cattle kill all the young trees, so preventing forest regenera- 
tion, and by their trampling, grazing, and other activities lead to 
the replacement of the characteristic forest-floor litter and vegeta- 
tion by Grasses (Fig. 86). For the Grasses are hemicryptophytes, 
having their buds protected within (or at least lying at) the surface 
of the soil, and far from being killed by grazing, may actually be 
stimulated bv it. Certainly they are encouraged by the removal of 
their broad-leafed and woody competitors, so that they will normally 
extend their area if the factor or factors suppressing these competitors 
are increased in intensity. Overgrazing may, however, lead to 
replacement of the Grasses by unpalatable herbs such as the weeds 
of open soils — perhaps followed by erosion if the rainfall is heavy, 
so that a barren waste may result (Fig. 87). 

These last biotic effects are commonly regulated by Man and tend 
to be destructive, often getting far out of hand, as in bad cases of 
erosion. However, animals are still often helpful to vegetation — 
for example in distributing seeds, fruits, and spores, etc., in effecting 
pollination, in loosening or compacting soil as well as in manuring 
it, in trampling seeds into the soil (for example, of range Grasses) 
and favouring regeneration, in keeping down injurious Rodents etc. 
(in the case of carnivorous Mammals and predatory Birds), and, 
in the instance of Man, in such activities as irrigation, the construc- 
tion of wind-breaks, soil-improvement and cultivation of many sorts, 
and the transplantation of useful plants and even the creation of 
new ones. 

Human activity is, indeed, the outstanding biotic factor in the 
world today, at least if we include consideration of Man's domestic 
animals. Especially is Man the enemy of forests, whether he 
realizes it or not. His ' shifting cultivation ' in the tropics is 
particularly damaging to vegetation, trees being girdled and the 
forest burnt, after which the accumulated fertility of the soil is 
exhausted in a very few years by cultivation ; thereupon a move 
is made, similarly to cultivate and desecrate a fresh area. Con- 
sequently some secondary and usually inferior type of forest now 
replaces the original one over vast areas. In other cases still more 
devastating erosion may result from forest clearance as indicated 
above, from over-pasturing, or from ' exsiccation ' leading to exten- 
sion of desert areas. In such instances of penetrating and serious 


Fig. <S6. — Some important effects of grazing. A, Sugar Maple-Beech forest in 
central Indiana, showing plentiful regeneration and litter; B, similar forest type 
in same region but subjected to heavy grazing by cattle. In the latter instance 
Grasses and a few unpalatable herbs have taken the place of the usual plants and 
litter of the forest floor, while tree regeneration has been eliminated so that the 
forest's life is jeopardized. (Reprinted with permission from R. F. Daubenmire, 
Plants and Environment, copyright date 1947, John Wiley & Sons, Inc.) 



disturbance, return to anything like the original vegetation is 
problematical. Appropriate measures following proper study are, 
however, nowadays leading to more and more effective conservation 
— for example of vegetation with the object of maintaining water 
supplies, of ranges against erosion, and of forests against fire. 
Elsewhere by irrigation or drainage, damming, building, persistent 
mowing, road and railway construction, mining and all manner of 
other enterprises, Man alters the water and other conditions over 

Fig. 87. — Devastating results of overgrazing. Rill, gully, and sheet erosion on 
overgrazed slope in San Joaquin Valley, California. (Courtesy of U.S. Forest 


vast areas — as he sometimes does the biological balance by tne 
introduction or extermination of various animals and weeds or other 

In many regions fire is an important factor which, because it is 
nowadays usually caused and controlled by Man, seems best con- 
sidered here. The effect is rather like grazing or cutting in that 
material is removed and if only small areas are affected they generally 
return to something like the former state — especially if the fire is 
a surface one that does not kill the larger thick-barked trees. How- 
ever, big and severe ' crown ' fires may destroy the humus and 
disseminules as well as most or all plants, and result in a community 
of colonists followed by a distinct ' burn succession ' of vegetation. 


Repeated burnings usually ehange the dominants and entire com- 
munities to ones which can regenerate after fires and, so to speak, 
withstand them. A few species, for example among the Grasses, 
seem actually to be stimulated by exposure to fire, while some 
Conifers, besides possessing thick and resistant bark, have cones 
which are opened by fire to liberate the ripe seeds earlier than would 
otherwise be the case. 

Although it may leave survivors open to fungous and other attack 
through unhealed burn-scars, fire tends to favour the more resistant 
species by removing their less resistant competitors ; it also alters 
many factors of the environment that favour some species rather 
than others. These principles apply to most types of vegetation, 
including forests, heaths, and grasslands. And whereas the general 
tendency of fire is towards vegetational degradation, the result is of 
course influenced by climate and human agency quite apart from 
the frequency and intensity of the burning. Indeed there are some 
circumstances in which firing will benefit desirable species and 
consequently may be intentionally practised by Man, as in pastures 
in many parts of the world where the old growth is burned off 

In conclusion we should outline the principles of cultivation, 
through which Man now largely controls the plant life of much of 
the land surface of the world and of some shallow waters. Cultiva- 
tion consists basically in preparing the surfaces of suitable substrata 
for the reception of seed (using this term in the widest sense), 
normally after the area has been cleared of any natural vegetation. 
Accordingly the early stages of growth of the crop are favoured bv 
a minimum of competition, although this may in time develop with 
weeds or between the plants of the crop. Consequently effective 
cultivation will include proper control of weeds and also sowing at 
such intervals of space as will reduce the ill-effect of competition 
to a minimum. Effort may also have to be given to maintaining 
other conditions favouring the growth of the crop. 

In order to continue satisfactory cropping year after year in the 
same area, it is commonly necessary to repeat for each crop such 
a ' tillage ' operation as digging or ploughing, which turns over the 
soil and helps to maintain it in a suitable state. Lime or peat or 
clay may also be added to some soils to improve their texture. 
Often there has to be also some form of manuring, from animal or 
mineral or chemical sources, to maintain the fertility of the soil by 
replacing the necessary substances which are removed in cropping. 


Moreover, in such arid lands as Mesopotamia, whence western 
civilization apparently sprang, successful cultivation also involves 
the artificial supply of water by irrigation. Conversely in many 
excessively humid or waterlogged areas, drainage is necessary before 
most crops can be sown. 

As a particular crop will remove more and more of the same (often 
necessary) substances from the soil, and perhaps add more and more 
of the same undesirable ones, it is common either to leave land 
fallow (i.e. cropless) for one year out of every two to four, or else 
to practise a ' rotation ' of different crops grown successively on an 
area. In such cases more or different weeds will be fostered, with 
obvious plant-geographical implications. Indeed the practices of 
and incidental to cultivation, such as removing natural or semi- 
natural vegetation, establishment (however temporarily) of artificial 
vegetation in the form of crops, the introduction of weeds and diseases 
whether they are controlled or not, and the opening up of fresh 
areas for plant colonization and succession, have a continuous and 
very widespread effect on the distribution and luxuriance of flora 
and vegetation in the world, and, accordingly, on local landscape 
and amenities. With Man's predominant position in modern times, 
his biotic or, more precisely, ' anthropic ' influence has become 
widely overwhelming. 

Further Consideration 

Three small introductory books, treating inter alia the factors of the 
environment, are : 

Sir A. G. Tansley. Introduction to Plant Ecology (Allen & Unwin, 

London, pp. 1-260, 1947). 
William Leach. Plant Ecology, fourth edition (Methuen, London, 

pp. vii -f- 106, 1956). 
H. Drabble. Plant Ecology (Arnold, London, pp. 1-142, 1937). 

Further useful treatments, which are mostly more detailed but by no 
means uniform in their groupings and terminology, include : 

A. G. Tansley & T. F. Chipp. Aims and Methods in the Study of 
Vegetation (Crown Agents for the Colonies, London, pp. xvi + 383, 

J. Braun-Blanquet. Plant Sociology (McGraw-Hill, New York & 
London, translated, etc., pp. xviii + 439, 1932). 

J. E. Weaver & F. E. Clements. Plant Ecology, second edition (McGraw- 
Hill, New York & London, pp. xxii -f- 601, 1938). 


R. F. Daubenmire. Plants and Environment' (Wiley, New York, pp. 
xiii + 424, 1947). 

H. J. Oosting. The Study of Plant Communities, second edition (Free- 
man, San Francisco, Calif., pp. viii -f 440, 1956). 

W. B. McDougall. Plant Ecology, fourth edition (Kimpton, London, 
pp. 1-234, 1949). 
Most of the above books give references to appropriate specialist ones, 

e.g. on soils. 

The following may be found pertinent in particular connections : 

V. J. Chapman. An Introduction to the Study of Algae (Cambridge 
University Press, Cambridge, Eng., pp. x -f- 387, 1941). Deals 
inter alia with the conditions in fresh and salt waters — see also our 
Chapters XV and XVI. 

V. J. Chapman. Salt Marshes and Salt Deserts of the World (Leonard 
Hill, London, pp. xvi + 352 + index, in Press). Describes the often 
peculiar conditions introduced by excessive salinity. 

H. U. Sverdrup, M. W. Johnson, & R. H. Fleming. The Oceans : 
their Physics, Chemistry, and General Biology (Prentice-Hall, New 
York, pp. x -f 1087, 1946). 

Useful recent works on the general ecology of plants and animals 
considered together include : 

E. P. Odum. Fundamentals of Ecology (Saunders, Philadelphia & London, 

pp. xii + 384, 1953). 
George L. Clarke. Elements of Ecology (Wiley, New York, pp. xiv 

+ 534, 1954)- 
A. M. Woodbury. Principles of General Ecology (Blakiston, New York 

& Toronto, pp. viii -f 503, 1954). 

Chapter XI 


Whereas originally the term ' habitat ' referred simply to the place 
(locality or station) in which an organism or community lived, 
ecologists nowadays take it to mean rather the kind of place, involving 
the sum of effective conditions (operative influences) characterizing a 
particular type of area or inhabited by a particular species or com- 
munity. Thus as a scientific term it involves all the conditions 
affecting an individual or community that are incidental to the place 
in which that individual or community lives, and we have to dis- 
tinguish between the general habitat of a community and the partial 
habitats of its component species, etc. These partial habitats may 
vary greatly within the orbit of a single example of a single general 
habitat. Thus with a Beech forest growing on a calcareous soil 
under certain climatic conditions, there may be rather similar general 
factors of soil and climate under which the dominant trees flourish 
in different spots within the same area or even in different areas in 
the same region. Very different may be, for example, the ' micro- 
habitats ' of Mosses or Algae growing on their boles or branches, and 
of herbs upon the forest floor. In the present treatment of the main 
different types of plant habitats that are discernible, we shall have to 
confine ourselves largely to general terms in showing how the factors 
of the environment dealt with in the last chapter constitute the 
habitats of the various types of vegetation to be treated in the next 
five chapters. 

The habitat is thus made up of the many and various environmental 
factors having any kind of influence upon life within it, and them- 
selves interacting complicatedly. The sum of effective ecological 
conditions has many widely different manifestations which range 
for instance from dry land to open water : and indeed water con- 
ditions seem to provide the best criterion for the primary subdivision 
of habitats. But even as most biological pigeon-holing is notoriously 
imprecise, involving many an arbitrary cut-off across a more or less 
complete line of gradation, so is it with any of the main types of 



tends to be least where the vegetation forms least of a ' show ', and 
especially where plants fail to stabilize the surface, as in the case of 
many dunes and coastal or desert or high-arctic areas. Plants are 
again relatively impotent in places of drastic topography and con- 
sequently strong geodynamic influences and recurrent catastrophe, 
for here the physical forces of nature usually rule, rather than the 
vegetation, and our recognition tends to be of habitats rather than 
of their inhabitants. 

Deserts are areas where the water conditions are too unfavourable 
(in the sense that the drought is lastingly too severe) to allow the 
support of any extensive continuous development even of short 
Grasses or scrub. They cover wide areas of flattish or other topo- 
graphy and in a sense are simulated on a small scale by areas of 
porous sand, gravel, shingle, or rock, where arid conditions may 
prevail even in regions of plentiful precipitation. The so-called 
cold deserts are the high-polar and high-alpine regions where frozen 
conditions make water unavailable to plants during most of the 
year. Even where the precipitation is extremely small in these 
rigorous regions, as in some high-arctic areas, there is, however, 
usually plentiful water available from melting snow for fair plant 
growth in favourable situations, at least early in the growing-season ; 
moreover there is normally frozen ground-water not far below the 
surface, so the regional appellation of ' desert ' seems inappropriate. 

In passing, mention should also be made of the so-called ' aero- 
plankton ', consisting of spores, etc., which float freely and unharmed 
in the air (although they can scarcely be considered as normally 
living thus), the microscopic ' cryoplankton ' which really live in 
and on the surface layers of snow or ice (see Chapter XV), and the 
' edaphon ', the flora and fauna of the soil, w r hich actually forms a 
special habitat for numerous recognized soil organisms. 

Aquatic Habitats 

Even when we divide these into the two main groups of saline and 
freshwater habitats, there are left intermediate ' brackish ' ones 
which seem best considered with salt waters. The degree of 
salinity can, and frequently does, greatly affect the habitat and 
attendant community. Freshwater habitats, apart from the mar- 
ginal ones already considered, comprise those of lakes, tarns, and 
ponds where the waters are relatively static, and those of rivers and 
streams where they are more or less dynamic. But streams can 


be reduced to chains of pools in dry weather, and lakes can have 
considerable convection and wind-engendered currents as well as 
run-off streams, so that here again any distinction is not wholly 
valid. Major variations in salinity found in different seas, estuaries, 
salt-lakes, etc., and the light, temperature, tranquillity or shelter 
from disturbance, size and depth of the body of water, possibility 
of attachment, and content of dissolved substances, are all im- 
portant factors leading to the existence of whole series of different 
aquatic habitats. Further variable factors may be the * reaction ' 
(acidity, neutrality, or basicity), aeration, and seasonal or tidal or 
other changes in the level of the surface. 

Let us briefly consider examples of these and some other factors 
as they may affect aquatic habitats : further details are given in 
Chapters XV and XVI. Light, being essential for photosynthesis, 
severely limits to those depths to which a sufficiency penetrates the 
possibilities for normal plant development, while towards the lower 
limit at which photosynthesis is possible, this vital function is in- 
sufficiently active to sustain life unless it be of specially adapted 
organisms. Some Red Algae seem to be so adapted, for they can 
grow at depths of nearly 200 metres in exceptionally clear seas, while 
some phytoplanktonic organisms have been dredged from, and can 
apparently live at, fully 200 metres. The larger Brown Algae, on 
the other hand, do not seem to be able to grow at any such depths, 
while vascular plants in fresh water usually extend no deeper than 
10 metres even in the clearest lakes, and in shallower water form 
zones correlated with their light requirements. However, in the 
Mediterranean Sea one flowering plant, Posidonia, is reported to 
extend down to depths of 80 or even 100 metres. 

Of the other factors, temperature differences frequently have much 
the same effect in aquatic as in aerial habitats, although major bodies 
of water will act as reservoirs militating against rapid changes in 
temperature. Consequently, conditions in water tend to remain 
more ' even ' than in the air, with the result that aquatic organisms 
and communities are often surprisingly widespread. Shelter from 
wave or ice action, and tranquillity from currents and tides, is another 
important factor profoundly affecting the habitat and attendant 
vegetation, macroscopic (i.e. visible to the naked eye, as opposed 
to microscopic) plants often being limited to sheltered bays, etc. 
This is often bound up with the size and depth of the body of water, 
on which convection and wind-engendered or other currents 
frequently depend, as does the degree (if any) of freezing. But 




such matters of size, depth, and shelter also introduce factors of 
their own, including light and temperature variations and the question 
of whether rooted or otherwise attached plants can grow up sufficiently 
to perform all their vital functions. This also depends upon the 
possibility of rooting or other attachment, which is usually dependent 
upon a suitable ' bed ' that of course varies for different types of 

As regards the content of dissolved substances, this can range 
from ' ocean ' or even more extreme salinity down to varying degrees 

Fig. 88. — Margin of tropical oligotrophic lake, with steep rocky sides and rapidly 
deepening water, supporting few larger plants. The hill-top vegetation is a semi- 
arid savanna with prominent Acacias. Lake Tanganvika, E. Africa. (Phot. 

R. Ross.) 

in ' fresh ' water. Often in bodies of fresh water the acidity and 
especially the nutritive salt content are of key significance for the 
development of planktonic communities — at least, within particular 
temperature ranges. In this connection it is often useful to dis- 
tinguish three types of such bodies, of which the first may give rise 
to the others : (i) oligotrophic, of waters poor in dissolved minerals, 
typically with Desmids abundant but supporting at most a narrow 
zone of rooted higher plants because of a hard rocky bottom and 
rapidly deepening water (Fig. 88) ; (2) dystrophic, with waters also 
poor in nutrients but rich in humus and acidic in reaction, often 
coloured, containing Desmids and Bog-mosses ; and (3) eutropliic, 


distinguished quantitatively from the oligotrophic type in bein ; 
usually poorer in the numbers of species but richer in individuals 
and poor in humus though commonly silted and shallow. The 
eutrophic type is also relatively rich in combined nitrogen, phos- 
phorus, and often calcium, typically contains plentiful Blue-green 
Algae, and has a broad zone of rooted Pondweeds, etc., and a sur- 
rounding one of luxuriant reed-swamp (cf. Fig. 89). 

Fig. 89. — Lake of eutrophic type near Prout's Neck, Maine. It is silted and 
shallow, with floating-leaf plants outside the broad marginal reed-swamp dominated 

by tall Cattails. 

In addition, the reaction (or ' pH level ', whether acidic or basic) 
of a body of water is commonly important to many organisms, 
while lack of oxygen may be a limiting factor deep down in sheltered 
situations. Also often limiting are seasonal and other changes of 
level in shallow places. Any substantial tidal activities are especially 
significant, as may be the speed and flow of currents and the presence 
and particular powers of various living organisms. 

Even as the zoned vegetation of sea-shores indicates the existence 
of different habitats at different levels, e.g. above and below normal 
low-tide mark, so do zones exist at different depths around the 
margins of deep lakes. Moreover, seas (such as the Sargasso) and 
especially lakes of warm regions, may bear extensive macroscopic 


floating vegetation. Yet, in major bodies of water, far more exten- 
sively occupied plant habitats are usually provided by the surface 
waters where sufficient light penetrates for photosynthesis. Here 
develop various planktonic communities of free-floating or swimming 
organisms, the vast majority of which are microscopic. The habitat, 
and consequently the community, may vary greatly with climatic 
factors and the presence of solutes and suspensions in the water, the 
whole being often subject to marked seasonal fluctuations including 
exhaustion of nutrients when the population is around its ' peak '. 
In the deeper layers of water and on deep ocean or lake floors where 
light does not penetrate, there are still habitats — especially for 
saprophytic plants living on the ' rain ' of sinking bodies. Indeed 
it is here that Bacteria are often especially numerous. 


We have seen that the environmental and internal factors of living 
organisms have intricate and highly complex interrelationships, be- 
longing as they do to a plethora of variables and potentialities that 
may be set in motion by all manner of ' master ' forces. Yet it is 
only the ' thin shell ' of environment directly impinging on, or 
immediately adjacent to, the organism that is of primary causal 
significance to it. So we get what in effect are ' partial habitats ' 
(microhabitats), for example in areas of drastic relief or uneven 
ground, or indifferent situations in a forest or other gross and complex 
community. Consequently it is rare for the measurements recorded 
by meteorological instruments to be actually those of the conditions 
of the microhabitat affecting the growing plant or, more precisely, 
the growing part of the plant. 

Microclimates are really the ultimate multiple expression of the 
local climatic effect which is so commonly and variously engendered 
by physiographic change, and microhabitats are their environmental 
result, though they may often be based on edaphic or other vari- 
ations. Very commonly one factor will compensate for another so 
far as some plants are concerned, but not in a manner satisfactory 
to the requirements of other plants — so leading to a jumbling of 
local communities — and this effect may be extended to microhabitats. 
Moreover, an alteration in one factor may initiate whole series of 
adjustments in others, often having far-reaching consequences. 
Especially are such effects apt to be complex when concerned with 
groups of factors that are closely related to one another — such as 


light, heat, and moisture relations, which vary simultaneously with 
every change in the intensity of insolation. 

In such circumstances it is difficult or sometimes impossible to 
segregate individual factors experimentally. Instead of physical 
apparatus we may use ' phytometers ', which are standard plants or 
clumps of vegetation that have the advantage of integrating all effective 
factors of the environment and expressing the result in their own 
responses. They react only to changes that matter to the plant or 
plants, whose protoplasm has the power of making adjustments. 
But much as species are usually composed of more or less numerous 
biotypes, so may general habitats or even single examples of them 
be made up of biotopes, which are the ultimate expression of environ- 
mental variation, being defined as the smallest natural area of space 
that is characterized by a particular environment. The biotope (or 
1 ecological niche ' of some authors) is thus the primary' topographic ' 
unit used in habitat classification. The community of forms in- 
habiting it is termed a biocoenosis, and the biotopes having similar 
characters are united into larger divisions called biochores. Many of 
these are apt to be represented in any small area, so whole series of 
phytometers, covering at least the different microclimates, may be 
necessary to determine the impress of even a single example of a 
general habitat ; moreover they should be accompanied by batteries 
of instruments adequate for the establishment of quantitative re- 
lationships between stimulus and response. 

In addition, matters can change rapidly with time. Thus even 
under a canopy of vegetation, such items as the movement of leaves 
by the wind, the changing angle of the sun, and various effects of 
weather and season, cause variations in the movement of shadows and 
sun-flecks across the ground. This in turn causes drastic changes 
in the amount of light-energy received at a given point — on which, 
as we have already seen, much else may depend. Altogether it is 
not surprising that experimental ecology has become an exacting 
(though scarcely exact) science. A saving grace is the fact that, 
apart from the grosser direct effects of animals and drastic physical 
agents, the actual effects of environmental factors upon plants are 
resolvable into a few physical and chemical processes. Examples 
are those which underlie the influence of light on photosynthesis and 
growth, the effect of temperature on chemical changes in the plant 
body, the evaporating power of the air on the water in the plant, 
and the effect of the soil solution on the absorbing organs and other 
parts of the plant. 


Drastic microhabitat development may take place at different levels 
or other situations in a forest, or on different sides of a hillock or 
even pebble. Thus the conditions under which an Alga or Moss 
lives on the bole of a tree are substantially different from those of 
an epiphyte in a crutch or on a branch high up in the crown, or of 
course from those of a herb on the sheltered forest floor. And again, 
the shelter from wind and sun given by even a minor projection from 
the ground, may enable a delicate plant or small community to grow 
there which could not exist in the exposed surrounding areas. The 
effect may even extend to the soils and their biota, and affect the 
plant habitat through them. Especially striking are the differences 
of temperature (and consequently of important dependent factors) 
on the north- and south-facing sides of tussocks at high latitudes in 
summer, which may vary by more than 20 C. within a few centi- 
metres, and allow active growth to take place in one spot when 
adjacent areas are frozen solid. Such effects may be the key not only 
to the micro-distributions of plants but also to their ranges over wide 
areas. Consequently it is important that we recognize the concept 
of microhabitat, for it is a very real and indeed fundamental one. 

Finally it should be recalled that not only different (phylogenetic) 
strains and even differently treated individuals may respond dif- 
ferently to microhabitat or other vagaries, but that different stages 
in the (ontogenetic) development of an individual may have critically 
different reactions, young seedlings being in general relatively feeble. 
Similarly, the tender young parts of older plants often differ greatly, 
in their resistance, from the remaining portions of the same plants — 
hence the familiar ' killing back ' of shoots by frosts in temperate 
regions, the rest of the plant being commonly unharmed. 

Main Successions 

When dealing in the last chapter with environmental factors we 
referred briefly to the competitive and other ' internal ' ones en- 
gendered by the plants themselves. Thus weeds compete for space 
and nutrients, some of those introduced to inhabited regions creating 
major nuisances by choking waterways, destroying the habitats of 
wildlife, or colonizing and rendering practically useless whole areas 
of agricultural land. The shade cast by dominant species, and the 
shelter they give, affect all the plants within the community ; also 
affected are the local atmospheric humidity and, often, soil structure 
and development as well as composition. 


It is a commonplace that units of vegetation, left to themselves, 
tend to change in a particular direction — usually from less complex 
communities of small plants to more complex ones dominated by 
larger plants of higher life-form (or, at all events, greater competition- 
impress). The change is continuous, recognizable ' stages ' being 
mere nodes of vegetational expression. Such is succession, the de- 
velopmental series of communities constituting a sere and leading 
up to a state of relative stability and permanence known as the climax. 
It should, however, here be admitted that not all ecologists accept 
the idea that vegetation can be widely interpreted in terms of develop- 
ment and equilibrium, while some, such as Professor Hugh M. Raup 
(in litt.), seem to doubt the validity of some of the basic assumptions 
involved — at least for those parts of the world in which they have 
themselves worked. Certainly, many of the beliefs involved are 
mere presumptions, or true only in some degree : thus successions 
may proceed only in relation to preceding and following stages, and 
climaxes are only relatively stable. This is often expressed by saying 
they are in ' dynamic equilibrium '. Nor is it for us to write into 
Nature's book meanings which she does not intend, or to attempt 
to inculcate for our own convenience an orderliness of pattern which 
does not exist. But if we deny the existence of seres and climaxes, 
we do away with two of the most stimulating concepts and useful 
tools of our trade, and so with this reservation it seems best to pro- 
ceed to use them. In doing so we ought also to bear in mind that 
many of the principles with which we are concerned have emerged 
from work carried out in temperate regions, and that in the Arctic 
(for example owing to frost action) and in the tropics (where there 
is often no clear dominance) things may be very different. 

We shall deal a little later with the typical stages of some charac- 
teristic seres, and, in the next section, with the main types of climax. 
With the reservations expressed in the last paragraph, some under- 
standing of these and allied concepts seems essential for an appreci- 
ation of the mosaic which is vegetation, and whose study, at least 
in terms of distribution, is the mainstay of modern plant geography. 
But first we should outline the component (often more or less con- 
tinuous) actions of a sere, which normally may be considered as 
follows : (1) nudation (the production of a bare area) is the initial 
prerequisite, whether it be by emergence or submergence, glacial 
recession, erosion, deposit, climatic change, or biotic agency. There- 
after follow (2) plant migration (including initial colonization) ; (3) 
ecesis (successful establishment) ; (4) aggregation of germules to 


form families (of a single species) or colonies '(of two or more species) ; 
(5) competition (virtually the struggle for existence) among the 
colonists, particularly for space, light, water, and nutrients ; (6) 
invasion by other plants, usually from adjacent areas ; (7) reaction, 
which essentially comprises the changes wrought in habitat 
conditions by the plants themselves {e.g. in soil formation) ; (8) 
coaction, the influence of organisms upon each other ; (9) stabiliza- 
tion, which of course is only relative, change being inevitable in all 
living organisms and aggregates thereof; and (10) attainment of a 
climax, by which time competition has generally become so intense 
that further invasion is problematical unless the community is 
drastically disturbed. 

The complete sere just indicated is a primary sere (prisere), be- 
ginning on a bare substrate without organic material. The chief 
types of primary seres are those initiated (1) in fresh water (hydro- 
seres), from which may be distinquished ' haloseres ' beginning in 
saline water ; (2) on damp aerial surfaces such as alluvial mud 
(mesoseres) ; and (3) on dry materials (xeroseres), of which out- 
standing examples are those starting on bare rock (lithoseres) and 
on dry sand (psammoseres). Secondary seres (subseres) are merelv 
partial, beginning after the succession has been stoppedj and thus 
not going back to a purely inorganic substratum unaffected by plants. 
They are distinguished as ' hydrarch ', ' mesarch ', or ' xerarch ', 
according to whether their initiation is under damp, median, or 
dry conditions, respectively. 

The broad tendency of succession is from simplicity to complexity 
of organization, and from dominance by lower to higher life-forms 
which make more and more exacting demands on the habitat. Yet 
sometimes we see ' retrogression ' to dominance by a lower life-form, 
for example when the habitat undergoes a change to less favourable 
water conditions. An incidental change in normal successions is 
from open to closed conditions, involving also an increase in the 
intensity of competition and marked alteration of local climatic and 
edaphic factors such as atmospheric humidity, wind, and the humous 
content of the soil. Such ' reactions ' are reciprocal, the plants 
affecting the habitat, which in turn affects the plants ; indeed, many 
of the higher life-forms only enter when the ground has been suitably 
' prepared ' for them by the dominants of earlier stages, which in 
turn they ruthlessly overshadow and frequently oust. 

We will now outline examples of the main types of priseres as 
postulated for temperate forested regions ; although the tendencies 


are generally similar, the outcome is apt to be different in polar and 
tropical regions, seres in the former being often much mixed and 
disturbed, and, in the latter, commonly retarded by lack of humus 
accumulation. These items are explained in the appropriate 

In temperate regions the typical hydrosere, after various non- 
essential (proseral) stages of plankton, etc., starts in bodies of fresh 
water whose beds w T here suitable are colonized by attached or other 
benthic (i.e. bottom) aquatic vascular plants and Mosses besides Algae 
as deep down as light conditions allow. These plants often form 
dense mats that collect silt and humus, there being frequently 
insufficient oxygen for rapid decay. The bed is thus built up 
gradually until, at a depth of some 1 to 3 metres, it can be invaded 
by floating-leaf types such as Water-lilies (Nymphaea spp.) or certain 
Pondweeds (Potamogeton spp.), which tend to shade out the sub- 
merged plants. The long stalks of these floating-leaf plants trap 
silt and their coarse bodies after death become deposited as, ulti- 
matelv, humus — so that the bed is built up with relative rapidity until 
the water is shallow enough for swamp plants to enter the community. 
These tvpically form a reed-swamp whose dominants are only partly 
submerged, building up the beds quickly and ousting the previous 
types. As the level continues to rise owing to the deposition of 
humus, etc., fully terrestrial invaders enter to characterize the sedge- 
meadow stage, the reed-swamp plants disappearing in due course 
as conditions are rendered unsuitable for them. With further rising 
in level of the soil surface and relative depression of the water-table, 
shrubs and ultimately trees enter and in time give rise to a hygro- 
phytic woodland. The Alders, Poplars, Willows, etc., which com- 
monly constitute this, will, in their turn, shade out the lower types 
and prepare for the climax forest which requires drier and more 
favourable soil conditions. Fig. 90 gives a diagrammatic repre- 
sentation of the stages of a hydrosere in section, a typical example, 
extending from the floating-leaf stage, having been shown in Fig. 89, 1 
while Fig. 91 continues this to the sedge-meadow and early timbered 
stages. A more detailed account of the early stages of some hydro- 
seres is given in Chapter XV. 

As a characteristic xerosere we will take a lithosere initiated on 
bare rock. Such surfaces are apt to be extremely difficult to colonize 

1 This was of eutrophic (' good foods ') type, an example of the oligotropic 
(' few foods ', geologically young) type, which may be expected in time to develop 
into a eutrophic lake, being shown in Fig. 88. 


and consequently may long remain uninvaded unless it be by pro- 
seral colonies of Bacteria, Blue-green Algae, etc. In time, however, 
the extreme exposure and general lack of water and nutrients are 
usually overcome by crustaceous Lichens or other hardy cryptogams 
as the first essential stage. They spread over the surface, helping 
the weathering forces of nature by corrosively or otherwise ' eating 
into ' the rock and adding plant material to form something of a 
nidus (nest) for ecesis of foliose Lichens, etc. These, attached at 

Fig. 90. — Diagram illustrating stages of hydrosere with deposition of peat. The 

usual stages are discernible on the gently-sloping shore, being, from left to right, 

submerged benthic, floating-leaf, reed-swamp, sedge-meadow (fen), hydrophytic 

scrub, hygrophytic forest, and finally climax forest. 


•? £)& *& .: » 


'£#■* r .-■/■ TM 

*///\"~ v -%r\?i;*' 

v ,%■ 

..-.«-. ■ . \ t 

Fig. 91. — Sedge-meadow stage of hydrosere colonized by some hygrophytic shrubs 

and, on right, trees, near Prout's Neck, Maine. An extensive reed-swamp is 

visible in the middle-distance, surrounding the lake shown in Fig. 89. 




a single point, overshadow the crustaceous types, holding water and 
fragments more effectively than their predecessors, and preparing 
the way for the moss stage whose components usually start entering 
as soon as soil particles accumulate in crevices and depressions. 
Such hardy Mosses are able to withstand prolonged desiccation and 
sometimes pioneer on uncolonized rock surfaces, making the Lichens 
unnecessary and ' proseral '. The Mosses usually enter as spores, 
and, among their closely aggregated axes and often densely matted 
rhizoids, young soil accumulates rapidly. Sometimes in the larger 
cushions this accumulation becomes quite thick (Fig. 92), forming 

Fig. 92. — Telescoped stages of xerosere in Norwegian Lapland. Gnarled speci- 
mens of the dominant Scots Pine have managed to grow in crevices of the rock 
whose surface is often still colonized by crustaceous Lichens, although in other 
places cushions of Mosses or ground-shrubs have accumulated some soil. 

fine nests for the colonization and establishment of herbs — especially 
annuals of xerophytic tendency. These contribute further to the 
humus accumulation and soil-building, and are typically followed by 
more exacting biennials and herbaceous perennials, which in their 
turn replace members of former stages while accelerating the further 
processes of succession. Often more important than colonization 
of the open rock surfaces is extension from crevices, at which Mosses 
and ground-shrubs are particularly adept. In time taller woody 
plants enter, constituting a less enduring stage that tends to overtop 
and oust the herbs but meanwhile to improve the soil and often 


conserve moisture, so that, in due course, forest and ultimately some 
kind of climax can develop (see below). 

The psammosere usually proceeds much more quickly than the 
lithosere, the initial problem being the ' binding ' of the surface sand. 
This is often accomplished by coarse Grasses or other ' advanced ' 
types (Fig. 93), while on gravel slides and talus slopes the pioneers 
may be coarse herbs or even woody plants, and succession still more 
rapid provided a reasonable degree of stability can be attained. 
Often, and especially on damp substrata that are immediately suitable 

Fig. 93. — Psammosere at Prout's Neck, Maine, showing Marram Grass (Attnno- 

phila arenaria) binding sand above high-tide mark. Some stabilized dunes and 

coniferous forest are seen behind. 

for colonization by advanced types, the phases of succession may be 
telescoped more or less completely ; but still the general tendency 
is evident. 

It may be noted that in these seres there is a general convergence 
of water conditions, the hydrosere becoming progressively drier and 
the xerosere progressively moister — until a mean is reached that in 
any given climatic region is approximately the same in the two cases. 
Typically this mean is inhabited by mesophvtes and is said to be 
' mesic ', though relatively xeric (dry) and hydric (damp) exceptions 
exist. It has been suggested that ultimately this mean should be 
the same under particular climatic conditions whatever the initial 
situation, but although this idea may be theoretically attractive it is 


evident that, at least in the present state of the world, different areas 
in the same climatic belt can support widely different climaxes. 

Main Climaxes 

Although competition is the chief key to succession, the final out- 
come of this latter lies in the population besi fitted (among those 
naturally attainable locally) to take advantage of the relatively mesic 
conditions brought about by past reactions. This population is the 
' climax ' (often more cautiously termed ' climax type ', and compare 
the reservation on page 323) and, being in close harmony with an 
essentially stable environment, is more or less permanent. Though 
by no means invariable in time and space, it shows a regularity of 
phvsiognomv and floristic composition that is usually lacking in 
successional stages. Thus in over-all form it persists as long as the 
climate remains unchanged — provided no new dominant enters or 
retrogressive change sets in, e.g. through impoverishment of the soil 
or accumulation of toxic substances. Dominance is due primarily 
to control of some ' key ' factor or factors of the environment, the 
dominant or co-dominants making of all plants present the greatest 
demands on the habitat, and normally when the climax is reached 
excluding invasion by any serious rival. Fluctuations in the com- 
munity thereafter tend to be minor in the absence of any forceful 

The climax is thus an equilibrated state of community compo- 
sition and productivity, that is adapted to maximum utilization of 
local resources by plants and, normally, animals. This maximum 
utilization is sustained, the climax being self-maintaining, and its 
efficiency is determined by the particular habitat as well as by the 
average climax population, which in turn is determined by migra- 
tional possibilities on one hand and, on the other, by all the factors 
that make up the mature ecosystem. It is evident, however, that, 
with changes in the environment, plant populations change from 
one type of area to the next, the vegetation varying largely according 
to local habitat. That was already indicated above. Consequently, 
climax and allied vegetation forms a pattern of communities varying 
with, and largely corresponding to, the pattern of environmental 
differences and gradients. 

In a general way climate determines the dominants and associates 
that can be present, and their life-form in turn characterizes the 
climax. This is really the mature stage of vegetation living in a 


state of more or less dynamic equilibrium with the local environment, 
though minor adjustments go on all the time. For life, as we have 
seen, can never be static, and the climax is only relatively so when 
compared with other stages of the succession. So besides the obvious 
differences in space, which are often attributable to diversification 
of the habitat, the climax inevitably shows some variation with time, 
its state remaining dynamic to that extent. This variation may be 
no more than that which results from the death and decay of indi- 
viduals especially of the dominant species — the disappearance of a 
big tree, for example, leaving a gaping hole in the forest canopy- 
followed by replenishment. If, on the other hand, there is a pro- 
gressive change, then we will have a continuing succession. 

The climax must at least be sufficiently stable and lasting to out- 
live the life-span of the dominant species. It commonly consists 
of patches or phases of different but related composition. However, 
these are normally at most representative of cyclic changes comprising 
upgrade and downgrade parts that nevertheless return to much the 
same climax type. If they do not do so,-then a succession or retro- 
gression must be involved, examples of the latter being the coloniz- 
ation of eroded heaths by Lichens and of coniferous forests by Bog- 
mosses. It is supposed by some that even these changes represent 
parts of a long-term cycle, but for such generalization the evidence 
seems inconclusive. We cannot wait long enough to see the true 
situation, which might take millennia to emerge ! 

In nature we expect to see some kind of ' regional ' or ' prevailing ' 
climax developed in local-climatically suitable situations at least on 
undisturbed tracts of the better soils of a region. But besides the 
complications already mentioned, there may be more important local 
variations of soil, biota, treatment, and so forth, causing the sere to 
be arrested, after being deflected, at some stage before the climax, 
and so constituting a subclimax. This is an imperfect stage in which 
the dominants are of lower life-form or competition-impress than 
those of the climax, the vegetation being ' held back ' by artificial 
or natural causes other than the climate. For the immediate site 
or ecological peculiarity may largely determine the actual growth. 
Examples are the subclimaxes due to such treatments as persistent 
burning or grazing (often called disclimaxes, being due to disturbance, 
or plagioclimaxes, owing to the deflection involved), or to marked 
differences in the rocky or other substratum. This last instance may 
if desired be termed an ' edaphic climax ' by those who doubt whether 
it will ever attain (or even if there is such a thing as) a ' regional 


climatic climax \ Similarly the biotically engendered disclimaxes 
may be termed ' biotic climaxes ', and those due to topographic 
features 'physiographic climaxes \ However, with removal of such 
a ' master factor ' as burning or grazing, the succession usually re- 
turns gradually to the autogenic main sere — at all events approxi- 
mately. And then there are instances in which an apparent climax 
constitutes in reality a preclimax in that its dominant or co-dominants 
are replaceable by one stage of others more advanced in life-form 
or competition-impress. This is really only a final type of sereclimax, 
or community arrested and held at some relatively early stage, and 
comprising the other form of subclimax with serai relationships 
simpler than the disclimax. Again by some authorities the term 
preclimax is used for one type of what we have here called sub- 
climax, namely that developing under locally unfavourable con- 
ditions. Finally there is the postc Umax, of vegetation more advanced 
than that of surrounding climax tracts, due to locally more favourable 
conditions obtaining in its limited area. This should be distin- 
guished from the relict community or fragment of a community 
that has survived some important change, whether this was toward? 
the improvement or detriment of the general environment. 

Whether or not we believe in the ' monoclimax ' hypothesis of a 
full regional climatic climax as not only the highest type which can 
exist in a given climate but also as the one which will ultimately 
develop more or less throughout the land area of that climate, the 
concept has its attractions and adherents. It appears, however, that 
soil (including water) and other conditions often prevent locally 
such ' regional ' development more or less permanently, and certainly 
no observer can wait to the end of the geological age and consequent 
termination of such an ' experiment '. Even if a general regional 
climatic climax were developed, who is to say that, for example, 
subsequent leaching or other effects might not lead to differential 
retrogression. Consequently it seems best to admit the likelihood, 
in any one region, of several different climax communities as repre- 
senting what may then be termed a ' polyclimax \ Even examples 
of the components of such a regional mosaic themselves frequently 
vary from spot to spot as well as with time, for example as indi- 
viduals in the dominant layer come and go and all manner of very 
local fragmentary seres are engendered ; indeed such variations can 
be so endless and perplexing that some ecologists, as has already 
been stated, doubt the validity of the very concept of climax, let 
alone its regional expression. 


What may well be the true situation is expressed by Professor 
H. J. Oosting, in the work cited at the end of the last chapter, as 
follows : 

' The Clementsian interpretation postulates that a climatic region has 
but one potential climax ; the most mesophytic community that the 
climate can support. It will be found on sites with average or in- 
termediate environmental conditions, particularly regarding moisture 
relations . . . Given sufficient time, with accompanying stability of 
climate and land surfaces, succession will have proceeded to such 
terminal, relatively mesophytic communities over much of the area. 
The stands will not be identical, yet they will have such a high degree 
of similarity that they are obviously related . . . The concept of a 
regional or climatically controlled climax necessarily includes recogni- 
tion of the convergence of successional trends toward a similar end. 
In its simplest statement, it implies that any habitat in a region, given 
enough time, could ultimately support a community representative of 
the formation. From this statement it might be inferred that a region 
of fairly uniform climate would eventually have a continuous and equally 
uniform vegetational cover throughout. Actually, this is never true 
. . . Locally, there are always edaphic or physiographic situations 
whose complex of environmental factors differ to such a marked degree 
from those of the general climate that they cannot support the regional 
vegetation type and probably never will . . . Monoclimax theory does 
not ignore these extreme situations but rather emphasizes that they are 
to be expected.' 

In any case the main, apparently climax, vegetational types of the 
world are regional and climatic to the extent that they tend to recur 
on favourable soils more or less throughout a region of particular 
climate ; they are also apt to have their counterparts in regions of 
different climate. These main types of vegetation are characterized 
by the life-form of the dominant or co-dominants and include : 
(i) tropical rain forests, the most luxuriant vegetation of all ; (2) 
tropical forest with a seasonal rhythm, due for example to monsoons ; 
(3) sclerophyllous forest, developed where there is a hot dry season 
and a cooler moist one, often merging into various parklands and 
savannas, which appear to belong rather with grasslands ; (4) warm- 
temperate rain forests, of evergreens, where there are few if any 
frosts ; (5) deciduous summer forest, with dominants losing their 
leaves in winter ; (6) northern coniferous forests, dominated mostly 
by evergreens ; (7) heath, dominated by members of the Heath 
family or heath-like plants such as Crowberry ; (8) tundra, the very 
variable but more or less continuous, treeless vegetation tvpical of 


many arctic and alpine regions ; (9) discontinuous ' fell-fields ' and 
sparser ' barrens ', etc., characterizing still more frigid regions ; (10) 
grassland, of various types dominated by Grasses and grass-like 
plants such as Sedges, often with scattered trees or shrubs forming 
a savanna; (n) semi-desert scrub ; (12) desert, with scanty but 
characteristic vegetation ; (13) mangrove ; (14) salt-marsh, which 
like some mangrove seems capable of persisting in the absence of 
disturbance ; (15) benthos, of submerged bottom aquatics ; (16) 
plankton of free-floating Algae, etc., including those of snow and 
ice ; and (17) the edaphon or soil communities including numerous 
Algae, Fungi, and Bacteria. Most of these are major, climatically 
determined vegetation-types [formations) of each of which various 
different ' aspects ' exist. Several are, however, apt to be serai in 
some instances — as are, of course, the many recognized stages in 
successions, such as the moss stage in the lithosere and the reed- 
swamp and bog or fen stages in the hydrosere. But strictly speaking 
a formation should represent the local climax. Examples of most 
of the above types and of some other (usually serai) ones, such as 
various swamps and marshes, are described (and often illustrated) 
in the next five chapters, which deal with the outstanding vegetational 
features of the world. 

Before describing the vegetational types of different regions and 
media, we must outline, in descending order of ecological status, the 
main classificatory units (eca) of vegetation which it seems practicable 
to recognize : 

(1) Formations. These are the great climatic units or regional 
climaxes such as desert, semi-desert scrub, tundra, deciduous forest, 
coniferous forest, broad-leafed evergreen forest, and some others, 
such as many heaths and grasslands which are determined par- 
ticularly by edaphic or biotic conditions but are so distinctive as to 
rank as formations. Each formation usually covers a wide area in- 
volving various conditions and so consists of more or less numerous 

(2) Associations. These are climax units dominated by normally 
more than one species having the life-form characterizing the for- 
mation to which their association belongs. An association exists 
under broadly uniform habitat conditions and is uniform in type so 
far as the general characters of the dominants and main associates 
are concerned. Such units become aggregated regionally to consti- 
tute formations. Examples of associations include various of the 
mixed deciduous forests of Old and New England, such as an 
Oak-Beech association. The developmental counterpart of the 


association is called an associes. It is a more or less advanced serai 
community dominated by more than one species, and is usually 
on its way to becoming an association. Commonly each association, 
having more than one dominant, is composed of two or more 

(3) Faciations (or else Consociations — see below). A faciation is 
a climax community with two or more, but less than the total number 
of, association?.! dominants. The serai counterpart of the faciation 
is the facies. Another local variant of the association is the lociation, 
which varies particularly in the composition of the important sub- 
dominants and influents. When there is only one dominant to 
each climax community we usually have 

(4) Consociations. These are smaller unit communities whose 
single dominant still has the life-form characterizing the formation. 
Such eca commonly occur on different soils, examples being the 
separate Oak and Beech consociations which make up the European 
Oak-Beech association. They may conveniently be named by adding 
-etum to the stem of the Latin name of the genus of the dominant, 
e.g. Quercetum (a consociation dominated by an Oak, Quercus) or 
Fagetum (dominated by a Beech, Fagus). The serai counterpart of 
a consociation, such as a reed-swamp dominated by a single species, 
is the consocies. Commonly recognized within a consociation or 
association are 

(5) Societies. These are minor (but still often apparently climax) 
communities that are commonly recognized within major eca, and 
usually owe their existence to local variations of habitat. They are 
dominated by one or more species other than the association domin- 
ants, and commonly are of lower life-form than these, being frequently 
subdominants of the higher econ, as in aspect (seasonal) and layer 
(stratal) societies. Thus a society represents a dominance within 
a dominance, whose dominant species is (or are) subordinate when 
we consider the association or consociation as a whole. Examples are 
the local and often very limited edaphic societies in many woodlands 
of temperate regions. The serai counterpart of the society is the 
socies, which, if it consists merely of two or more invading species 
without evident associates, may be called a colony. Within societies 
etc., there may be 

(6) Clans. These represent the lowest climax unit, consisting 
each of a small aggregation of a single very locally but overwhelmingly 
dominant species. The serai equivalent is the family, derived from 
the multiplication and gregarious growth of a single immigrant. 


Further Consideration 

Most of the subjects treated in this chapter will be found discussed 
— though often in a different light — in each of the books of Tansley, 
Leach, Drabble, Weaver & Clements, Oosting, and McDougall cited at 
the end of the preceding chapter. 

More specialized works dealing with the aspects indicated by their 
titles are : 

F. E. Clements. Plant Succession : an Analysis of the Development of 
Vegetation (Carnegie Institution of Washington, Publ. No. 242, 
pp. xiii -f- 512, 1916). 

F. E. Clements, J. E. W t eaver, & H. C. Hanson. Plant Competition : 
an Analysis of Community Functions (Carnegie Institution of Wash- 
ington, Publ. No. 398, pp. xvi + 34°> J 9 2 9)- 

Sir E. J. Russell. Soil Conditions and Plant Growth, eighth edition 
edited by E. W. Russell (Longmans, London etc., pp. xvi -f- 635, 

P. J. Kramer. Plant and Soil Water Relationships (McGraw-Hill, New 
York etc., pp. xiii -(- 347, 1949). 

R. Geiger. The Climate Near the Ground, translated by M. N. Stewart 
and others (Harvard University Press, Cambridge, Mass., pp. 
xxi -f- 482, 1950). 

It may be noted that whereas ecologists are notoriously prone to 
make and use technical terms (so that humorists say they will even 
call a spade a ' geotome '), of which not a few have been introduced 
in the above chapter, these latter were in most cases selected either 
for their precision-giving value or because they would be needed 
elsewhere in this work, and particularly in the following chapters 
dealing with the main vegetation-types of the world. Many of these 
terms occur again and again, though in other cases such non-com- 
mittal words as community (for any grouping of plants, or econ) 
or ' ecotone ' (denoting the transition zone between two communities) 
are employed. Several ecological terms, such as association and 
subclimax, are unfortunately liable to be used in entirely different 
senses by members of different schools of ecological thought ; for 
the present work the most widespread or generally appropriate use 
has been chosen, others being commonly ignored in the interests of 
simplicity. A definition or other indication of the sense employed 
has usually been given on introduction of each technical term in 
this book, and may be found through the index. 

Chapter XII 


We now come to what in some respects is our main objective — 
the study and interpretation of the vegetation and its component 
communities inhabiting different areas of the world. It should, 
however, be remembered throughout the following treatment that 
the various types of vegetation described are merely those which 
we recognize, almost all being apt to intergrade with little or no 
distinction or even characterization. These intergradings and also 
the relative positions of the main vegetational types will of necessity 
be largely ignored in the following brief treatment, although the 
geographical situation and main neighbouring types in each instance 
can be noted in a general way from the map facing page i of the 
text, which is a highly generalized vegetation map of the world. 

We have seen that the systematic relationships of the flora of a 
region depend to a considerable extent upon its geographical con- 
nections or barriers, whether past or present ; on the other hand 
the physical characteristics of the vegetation are largely conditioned 
by local environmental factors. Thus when two areas have been 
separated since far back in geological time by such barriers as wide 
seas which cannot ordinarily be crossed by plants, their floras (of 
component species, etc.) will often be very different, whereas if 
their environmental conditions are similar their vegetation in closely 
comparable habitats is likely to have the same general appearance. 
This is because similar external conditions which make up particular 
habitats tend to produce communities (and life-forms as regards 
component plants) whose external physical features are much alike 
—however dissimilar may be their more fundamental reproductive 
and allied structures by which we usually classifv them. For this 
reason we expect — and generally find — in hot arid habitats succulent 
plants belonging to very various families, in moist temperate regions 
deciduous trees, and on high mountains dwarf shrubs and perennial 
herbs of tussocky growth. Nor does it matter in this connection 
whether or not the areas concerned are separated from one another 



by thousands of miles, or, within reason, wherever in the world 
they may be. For whereas a single factor of the environment may 
result in a characteristic vegetational feature, it usually takes the 
entire habitat complex to stamp the community fully. Consequently, 
to the extent that distant habitats may be comparable in supporting 
similar vegetation-types, we may deal with them on a world basis. 
In this the first division is climatic, followed by local differentiation 
caused by edaphic or other differences : and as the vegetational 
types of temperate and adjacent lands are the most familiar to the 
greatest number of us, we may conveniently start with them. 

Deciduous Summer Forests 

Summer-green forests, dominated by broad-leafed trees which 
lose their leaves for the unfavourable period of winter, constitute 
the main climax formation over much of temperate Europe, eastern 
Asia, and North America, reappearing in some comparable regions 
of the southern hemisphere. From the physiological point of view 
the cold winter tends to be a dry period, owing to the fact that low 
temperatures often hinder absorption of water by the roots : this 
is counterbalanced by the leafless condition during winter, for it 
is chiefly from the leaves that loss of water takes place, and it is 
mainly such loss which has to be made good by absorption from 
the soil. If active transpiration continued when the resultant 
water-deficiency could not be made good by absorption, owing for 
example to warm weather when the soil remained frozen, serious 
injury and even death might result. As it is, these deciduous broad- 
leafed forests occupy many of the most populous regions of the 
world, and although in such areas we may now see around us only 
patches of anything approaching a climax, this situation is largely 
due to human disturbance — to clearance for agricultural or other 
purposes, or to the depredations of Man's domestic animals. For 
the conditions that have favoured the growth of such forests have 
also favoured the development of the most prosperous agriculture 
and grazing, including widespread cultivation of cereals, and con- 
sequently of some of the highest stages of human civilization. 
Owing to the deciduous habit of the main dominants and the char- 
acteristic dying down of many of the associated plants, these forests 
look entirely different in winter (Fig. 94) and summer (Fig. 95). 

Deciduous summer forests have their main development (1) in 
eastern North America in the temperate belt northwards to the 


Great Lakes and the upper reaches of the Gulf of St. Lawrence 
and westwards beyond the Mississippi ; (2) in temperate western 
Europe whence they extend eastwards to the Urals as a wedge 
between the northern coniferous forests and the southern steppes, 
reappearing in the Caucasus region ; (3) in northern Japan and 
adjacent parts of eastern continental Asia ; and (4) in the southern 

Fig. 94. — Leafless condition of mixed deciduous summer forest in northeastern 
United States — in winter. 

hemisphere in limited parts of Patagonia, southern Chile, and Tierra 
del Fuego. In all these regions there is a cool to severe winter 
but otherwise temperate and moist climate, with some precipitation 
all the year round, and a total of from 70 to 1 5 1 or more cm. (approxi- 
mately 28 to 60 inches) annually. 

In contrast to the situation in most tropical forests, the trees in 
deciduous summer forests form only a single main stratum or storey, 


though there may be tall shrubs or some smaller (often young or 
unsuccessful) trees forming a partial second stratum below. Such a 
lower stratum is especially to be seen when the main tree storey is 
not well developed, for when the latter is dense there is normally 
but scanty development of tall plants below it, and not much even 
of under-shrubs and herbs. There are also few climbers in most 
deciduous summer forests, while any epiphytes consist usually of 
lowly cryptogams. Consequently such temperate forests tend to 

Fig. 95. — Mixed deciduous forest in northeastern United States — similar to that 
shown in Fig. 94, but in summer. (Phot. G. E. Nichols.) 

be far less luxuriant than most tropical ones, as may be seen on 
comparing Figs. 94 and 95 with Figs. 143 and 144. Often there 
is far less undergrowth in the temperate forests than is shown in 
these photographs, so that the contrast with tropical rain forests is 
still more striking. 

In deciduous summer forests the winter buds burst forth and the 
leaves of the dominants expand quickly, soon after the growing- 
season starts with the advent of suitable temperatures. The foliage 
is thus fully developed quite early in the season, so that little time 
is lost. Flowering also tends to be completed early, giving ample 


time for the development and ripening of the fruit ; indeed in many 
types the flowers open before the leaves expand, this ' spring flower- 
ing habit ' allowing freer aeeess of wind and the early insects for 
pollination. Also often taking advantage of the brief period before 
the leaves of the dominants or other tall plants expand and cut off 
most of the sunlight, are the small- perennial herbs with persistent 
underground portions that send up flowering shoots and leaves very 
early in spring, so constituting the ' prevernal aspect '. They 
flower and fruit rapidly and often die down soon afterwards, as in 
he case of the Lesser Celandine (Ficaria verna) and Virginian Spring- 
beauty (Claytonia virginiana). Somewhat later are the plants of 
the ' vernal aspect ', such as the Wood-sorrel (Oxalis acetosella) and 
Yellow Archangel (Galeobdolon luteum), which flower during the 
bursting of the buds and expansion of the leaves of the overtopping 
types. Thus not only are the winter and summer aspects of such 
forests strikingly different, but also often the spring and autumn 
ones — for in this last season the brilliant galaxy of falling leaves, 
including the reds of Maples, the yellows of Birches, and the oranges 
of various others, makes it in the opinion of many enthusiasts the 
most beautiful of all. 

Although they naturally often intergrade as well as vary com- 
plicatedly, some five main types of deciduous summer forests may 
be recognized in various temperate parts of the world, as follows. 

i. Oakwoods of western and central Europe, which tend to be 
relatively open and light. The dominants are the Pedunculate Oak 
(Quercus robur) and/or the Sessile Oak (O. petraea = O. sessiliflora), 
the most important associated trees including Ash (Fraxinus excelsior), 
Poplars (Populus spp.), Birches (Betula spp.), Elms (Ulmus spp.), 
Alder (Alnus glutinosa), and Wild Cherry (Primus avium). These 
change largely according to the nature of the ground. Smaller 
trees and tall shrubs flourishing in the comparatively light shade 
include Hazel (Corylus avellana), Holly (Ilex aquifolium), Hawthorn 
(Crataegus monogyna), Eield Maple (Acer campestre), Crab Apple 
(Pyrus malus), Mountain-ash (Sorbus aucuparia), and Yew (Taxus 
baccata). In a still lower layer are found a great variety of under- 
shrubs and coarse herbs and Grasses, while cryptogamic epiphvtes 
may flourish on the bark of the trees. In areas of prevailingly high 
atmospheric humidity, some vascular plants may grow as more than 
fortuitous epiphytes. Ivy (Hedera helix) and Honeysuckle (Lonicera 
periclyv.enum) are common woody climbers in European oakwoods. 

2. The more varied and luxuriant mixed forests of eastern North 


America, eastern Asia, and southeastern Europe, which differ much 
from the above in systematic composition, but nevertheless tend to 
be roughly comparable with them in physical form. Thus various 
different Oaks (Ouercus spp.), Beeches (Fagus spp.), Birches (Betula 
spp.), Hickories (Carya spp.), Walnuts (Juglans spp.), Maples (Acer 
spp.), Basswoods (Tilia spp.), Elms (Ulmus spp.), Ashes (Fraxinus 
spp.), Tulip-trees (Liriodendron sp. or spp.), Sweet Chestnuts 
(Castanea spp.), Hornbeams (Carpinus spp.), and many others, here 
mav vie with Conifers such as Pines and Spruces and their allies 
— often doubtless as a result of disturbance. The undergrowth is 
commonly luxuriant and various, as is the herb layer especialily 
where light penetrates, while climbers are relatively plentiful. The 
range and variety, not only between the different regions but even 
within the main individual ones, is far too great for us to do justce 
to here, much less to describe the types in any detail. Those of 
eastern North America are well treated by Braun in the book cited 
at the end of this chapter, and those of eastern Asia are described 
by Wang Chi-wu in a work which it is hoped may soon be published 
by the successors of Chronica Botanica. Examples from eastern 
North America are shown in winter and summer aspects in Figs. 94 
and 95, respectively. Here the main pertinent types to be recognized 
include : (a) the mixed mesophytic type of moist but well-drained, 
unglaciated plateaus, e.g. of the Appalachians, in which dominance 
is shared by a number of species of trees, particularly American 
Beech (Fagus grandif olid), Tulip-tree (Liriodendron tulipifera), several 
kinds of Basswood (Tilia spp.), Sugar Maple (Acer saccharum), Red 
and White Oaks (Ouercus rubra s.l. and O. alba), and Hemlock (Tsuga 
canadensis) ; (b) the mixed Oak-Hickory type of southern mid- 
western uplands, extending northwards on to ' glaciated ' territory 
and southwards with the admixture of abundant Pines ; (c) the 
Oak-Chestnut type extending eastwards on to the coastal plain from 
northern Virginia northwards and, now that the Chestnut has largely 
disappeared as a tree owing to fungal ravage, dominated chiefly by 
White Oak, Red Oak, Chestnut Oak (Ouercus montana), and Tulip- 
tree ; (d) the Beech-Maple type lying in the glaciated territory to 
the north of the mixed mesophytic type, and dominated chiefly by 
Beech and Sugar Maple, though many areas are youthful and still 
serai ; and (e) the Maple-Basswood type centred on the driftless 
area of Wisconsin, in which Sugar Maple and Basswood (Tilia 
americana) are the dominants of the climax, which usually contains 
much associated Red Oak. In Japan and adjacent China, etc., much 


the same genera (apart from Hickory) but different species are 
usually involved in the deciduous forests, examples among the Ashes 
being Fraxinus mandshurica and among the Birches Betula ermanii, 
while the Beech Fagus crenata characterizes fine ' Buna ' forests. 

3. The Beech forests which, especially in Europe, where Fagas 
sylvatica is the species involved, form an almost uniform closed 
canopy, intercepting the sunlight so effectively that few shrubs and 
herbs can grow below. The trunks are tall and slender, particularly 
when the trees grow closely together, and few competitors of the 
dominant Beeches are able to enter their preserve. A thick brown 
mat of fallen leaves and leaf-mould covers the ground. It is chiefly 
in early spring before the Beech leaves expand that small perennial 
herbs such as Bluebell {Endymion (Scilla) non-scriptus) and Wood 
Anemone {Anemone nemorosa) tend to flourish as a prevernal aspect, 
often growing gregariously to form attractive carpets. Otherwise 
herbs are characteristically few, sometimes being largely limited 
to nongreen saprophytes, while the lower shrubs may consist of 
no more than scraggy Brambles {Rubus spp.) in the more open areas. 
Occasional Ash, Wild Cherry, White-beam (Sorbus aria s.l.), or 
other trees may reach the height of the canopy, or Hollies or Yews 
form a scraggy subordinate layer. The commonest tall shrubs are 
Elder {Sambucus nigra) and Field Maple, with Spindle-tree (Euony- 
mus europaeus), Dogwood (Cornus sanguinea), and Wayfaring Tree 
{Viburnum lantana) occurring chiefly in openings. 

4. Southern Beech (especially Nothofagus antarctica) forests, 
usually with associated evergreens such as Drimys zuinteri, of southern 
South America. The trees are closely crowded, shrubs again being 
relatively few. Ferns and Bryophytes, on the other hand, are 
numerous and often extremely luxuriant, the latter sometimes 
forming an almost continuous carpet over the ground and fallen 
logs, etc., and extending well up the standing trunks. 

5. The damper aspect of deciduous woodland developed especially 
on marshy ground subject to inundation, and dominated by various 
Alders {Alnus spp.), Willows (Salix spp.), Poplars, Birches, and the 
like, with often a tangle of hygrophytic shrubs. Climbers and 
epiphytes may be plentiful, and on the moist ground often flourish 
coarse herbs and tussocks of tall Grasses, Sedges, and Ferns such as 
the Royal Fern {Osmunda regain agg.). These marshy thickets are 
plentiful in temperate regions on both sides of the North Atlantic, 
and appear to represent late stages in the hydrosere. 

In addition, the Sweet Chestnut forests of various parts of southern 


and eastern Europe are sometimes described as a further type, and 
the drier, park-like woodland on limestone hills of central Europe 
as yet another. Deciduous ' parklands ' also occur in the northern 
prairies of North America, and though dominated by groups of Aspen 
and other Poplars (Populus spp.) in a manner reminiscent of serai 
stages, appear in Alberta and Saskatchewan to constitute ' a forest 
type in its own right ' (H.M. Raup in litt.). Also characteristic, if 
limited, are the deciduous forests of the Pacific coastal regions of 
the northern United States, and the very luxuriant ones of the 
western Caucasus where various kinds of Oaks, Beeches, Maples, 
Horse-chestnuts (Aesaihis spp.), Cherries, and Cherry-laurel (Prunus 
lanrocerasus) are commonly mixed with Conifers and a wealth of 
shrubs and climbers. On the other hand, the open Birch forests of 
northernmost Scandinavia, etc., in spite of their broad-leafed 
deciduous nature, belong to the next group. 

Northern Coniferous Forests 

These are also known as ' boreal forests ', ' subarctic forests ', or 
1 taiga ', although it seems preferable to reserve the last term for 
their open, park-like northern tracts (see pp. 346-8). 

The main dominants of these forests, instead of having broad 
leaves which they shed in winter, typically solve the problem of 
perennation through that unfavourable period by having narrow or 
small, needle-like or sometimes scale-like leaves. Besides their size 
and shape, these leaves usually have other xeromorphic characteristics 
that help to reduce transpiration to very modest rates. Consequently 
they can be retained in winter, most such trees being evergreen and 
having the advantages over deciduous types of being able to photo- 
synthesize whenever conditions allow, and meanwhile of saving the 
wastage involved in complete annual leaf-fall. However, it is as 
though at their northernmost extremity such trees were unable to 
support winter transpiration, for the Larches (Larix spp.) among 
these needle-leafed types are regularly deciduous, losing their leaves 
every autumn, and in places persisting farther north than the ever- 
green trees. Thus it is the Dahurian Larch (Larix dahurica) that 
alone forms the farthest north ' forest ' in the world (at about 
72 50' N. and 105 E. in Siberia). 

Although the main dominants of these hardiest of forests are 
needle-leafed Spruces, Pines, Firs, and other Conifers, they often 
have associated broad-leafed deciduous Birches, Poplars, and the 




like — indeed Birches actually form the northernmost (albeit scrubby 
and open, see Fig. 96) ' forests ' of much of Europe, and supply 
the only native arborescent growth in Iceland and Greenland. The 
groves of Aspen and Balsam Poplars (all Populus spp.) are usually 
(but apparently not always — see p. 343) secondary, replacing conifer- 
ous stands after felling or fires, while those of Alders (Alnus spp.) 
are normally merely serai. 

The undergrowth and ground-flora in well-developed examples 
of these coniferous, etc., forests tend to be less dense and diverse 

■ WiBik : mmm- 

Fig. 96. — Open scrubby Birch ' forest ' in Finmark, northern Norway (' Nor- 
wegian Lapland '). Dark ground-shrubs and Lichens cover the ground, the 
former being here locally predominant. 

than in most broad-leafed deciduous ones. The reasons are 
apparently (a) that there is no season during which the lower layers 
are not shaded by the leaves of the dominants, (b) that the thick 
and dry carpet of slowly-decaying resinous leaves hinders the 
establishment of seedlings, and (c) that the generally less favourable 
regions offer fewer potentialities for growth. Nevertheless there 
may be a fair shrub layer, especially in the damper situations where 
the ground is commonly moss-covered ; on the other hand, in 
drier areas and in the northernmost sparse forests, luxuriant Lichens 
often form a continuous ground-investment over vast areas. 

One or another type of northern coniferous forest (or, occasionally, 


its broad-leafed faciations of Birch or facies of Aspen) occupies 
most of the northernmost belt of forested terrain around the cool- 
temperate and boreal shoulder of the globe. Not only do they form 
the northern, but in many places also the altitudinal, limits of tree 
growth, extending southwards through often some 15-20 degrees of 
latitude from this northern limit, with outliers or tongues still 
farther south — for example in the eastern and western United States 
of America. There are also important outliers in the mountains 
of central and southern Europe and Asia (cf. Fig. 83, B). In addition, 
the Pines are highly developed in area as well as species in Mexico 
and throughout much of the Caribbean region, and, with other 
genera of Conifers, are important far south in Central America and 
the Mediterranean region, though these vegetation-types scarcely 
belong to the above series. For in general the most characteristic 
and extensive needle-leafed coniferous forests are developed chiefly 
between the 45th and 70th parallels of north latitude. Nor, in view 
of the very limited persistence of ice-free land at corresponding 
latitudes in the southern hemisphere, is it surprising that such forests 
are not paralleled there, the austral Conifers, although evergreen, 
being commonly broader-leafed and more warmth-loving (or at least 
far less cold-resistant). 

Except when deciduous trees or shrubs are plentiful, there is 
relatively little difference in the appearance of the northern coniferous 
forest at different seasons — apart, of course, from lying snow which 
characterizes much of this belt more or less throughout the winter. 
The climate is cool, with the winter extremely cold in the continental 
regions ; indeed most of the lowest temperatures ever recorded have 
been within this belt in the interior of the great northern continents. 
In view of this frigidity, the absolute humidity need not be high 
for the climate to remain relatively moist, with fairly regular pre- 
cipitation throughout the year. The soil is generally poor and of 
glacial origin. Ignoring most of the southern outliers, the following 
five main types of northern coniferous and allied forests may be 

1. The usually mixed coniferous forests occupying most of the 
boreal forested parts of Eurasia and North America, dominated by 
various assortments (or occasionally a single species) of Spruce, Fir 
(Abies spp.), Pine, and Larch. The precipitation is usually between 
25 and 100 cm. (approximately 10 and 40 inches) per annum and 
the mean temperature of the warmest month above io° C. (50 F.), 
though the summer is relatively short. In eastern North America 


and westwards to the Rocky Mountains the local dominants in the 
northern or ' Hudsonian ' belt are usually one or more of the follow- 
ing : White Spruce (Picea glauca agg.), Black Spruce (P. mariana), 
Tamarack (Larix laricina), Balsam Fir (Abies balsamea), Jack Pine 
(Pinus banksiana), with or without associated broad-leafed Poplars, 
Birches, or American Aspen (Populus tremuloides). Farther south, 
other dominants enter and in time there is a grading into the 
deciduous summer forest. Tall shrubs such as Willows (Salix spp.), 
Scrub-birches, and species of Dogwood (Cornus) and Pimbina etc. 
(Viburnum) may be plentiful, especially in damp situations where 
Mosses form a characteristic carpet. Herbs are, however, usually 
little in evidence, the normal ground-flora being heathy, including 
various species of Blueberry etc. (Vaccinium spp.), Labrador-tea 
(Ledum spp.), Pale-laurel (Kalmia spp.), and the Crowberry 
(Empetrum nigrum s.l.), typically alternating with or growing out of a 
lichen-rich carpet in the drier situations. In the damper situations 
the ground layer is contrastingly mossy. Farther west the Balsam 
Fir and Jack Pine are replaced by other species, particularly by 
Alpine Fir (Abies lasiocarpa) and Lodgepole Pine (Pinus contorta 
var. latifolia), while in Eurasia the dominant species are different 
again. Thus in northern Europe the Scots Pine (Pinus sylvestris) 
is often the sole dominant in the west, with Norway Spruce (Picea 
abies) entering to the south and east, and, farther east still, Siberian 
Spruce (Picea obovata), Siberian Larch (Larix sibirica s.l.), Siberian 
Fir (Abies sibirica), and, ultimately, Siberian Stone-pine (Pinus 
sibirica). Except for Norway Spruce these all persist at least well 
into Siberia proper, or are confined thereto. In the centre and east 
of Siberia, however, the Siberian Larch is replaced by the Dahurian 
Larch (Larix dahurica) and the Siberian Stone-pine by the Siberian 
Dwarf-pine (Pinus pumi la). Apart from this last, which is shrubby, 
most of these Conifers (including the Lapponian form of Scots 
Pine) are short-branched, at least above, to give a conical shape. 
Moreover they tend to be shallow-rooting, and consequently able to 
grow in open canopy in areas where the subsoil is permanently 
frozen. Examples are seen in Figs. 97 and 98, the ground-flora 
almost everywhere being of the characteristic heathy type. 

2. The open park-like ' taiga ' occurring towards the northern 
limit of arborescent growth. This is really only the product of 
depauperation of the various faciations of northern coniferous forest 
just described, but it is so striking in appearance as to warrant 
separate mention. It is characterized by the rather sparsely and 




Fig. 97. — Boreal coniferous forest surrounding lake-side bog in sheltered valley 
in Troms, far north of the Arctic Circle in Norway. The dominant is' Scots 
Pine of the conical Lapponian form. The ground-vegetation is heathy, but in 
drier situations is apt to consist largely of light-coloured Lichens as in Fig. 98. 

Fig. 98. — Outside the taiga in northern Ungava, Canada. The scattered domin- 
ants of the taiga, seen in the middle-distance, are Black Spruce (Picea mariana) 
and Tamarack (Larix laricina), the ground between, as in the foreground, being 
largely occupied by swarded Lichens, though many licHen-covered boulders and 
dark patches of Crowberry (Empetrum) and Mosses are visible. In centre is seen 
a typical Spruce outlier consisting mainly of a lower deck that is protected by 
snow in winter (cf. Fig. 82, A), but with some puny stems straggling above. 


often evenly spaced dominants, poverty in associated vascular plants 
(apart from xeromorphic ' heaths', such as Crowberry and species 
of Vacciniwn, in the shelter of the dominants), and richness of the 
lichen carpet at least in dry places (Fig. 98). The Lichens are 
usually intricately mixed and inclusive of so-called Reindeer-mosses 
(Cladonia spp.) and Iceland-mosses (Cetraria spp.), being typically 
aggregated into a pale but dense sward several centimetres thick. 
In damp depressions and especially along the courses of rivers, there 
occur faciations approaching the ordinary northern coniferous forest 
of the region. These may project as timbered tongues containing 
well-grown trees, or even form outliers in the tundra — so con- 
stituting the so-called ' forest-tundra \ Such better growth often 
seems to be correlated with better aeration of the roots where there 
is active drainage of water (as confirmed by Professor Harold J. 
Lutz, voce). In general, however, the dominants of the open 
' taiga ' are of poor development, often gnarled and only a few feet 
high though ancient, 1 conditions for growth being here largely 
unfavourable. These dominants are usually the one or more hardiest 
tree types of the forest lying to the south — for example, White or 
Black Spruce and/or Tamarack across most of northern Canada, 
Dahurian Larch in much of Siberia, and a scrubby Birch (Betula 
odorata) in northernmost Scandinavia (Fig. 96). 

3. The Pacific ' coast forest ' of western North America, developed 
chiefly from southern British Columbia to northern California, but 
with some of the main dominants extending much farther north- 
wards as well as southwards. The region is one of equable climate 
with high rainfall and atmospheric humidity, and supports the 
densest coniferous forest of the world as well as some of the biggest 
and tallest of all trees — e.g. Coastal Redwood (Sequoia sempervtrens), 
Big-tree (Sequoiadendron giganteum), and Douglas Fir (Pseudotsuga 
taxifolid). These may reach heights of around 100 metres, or, in 
the case of the first-named, no metres, with trunk girths often over 
20 metres in the first two cases. A forest of Coastal Redwood is 
shown in Fig. 18, D. Various further Conifers belonging to several 
different genera constitute the other main dominants, etc. Although 
Ferns, including the common Bracken (Pteridium aquilinum agg.) and 
Hard Fern (Blechnwn spicant), are widespread, herbs in general are 
little in evidence. However, many characteristic shrubs occur, 

1 Near the northern limit of forest in the Northwest Territories of Canada, 
the author has counted more than ioo growth-rings in Black Spruces with trunks 
barely three inches in diameter at the base and five feet high. 




including kinds of Spicy-wintergreen (Gaultheria), Barberry (Ber- 
beris), Currants etc. (Ribes), Rhododendron, Elder (Sambucus), and 
Blueberries etc. 

4. The so-called ' lake-forest ' of the eastern half of North 
America, lying between the general northern Hudsonian belt and 
the deciduous summer forest to the south, and centred on the 
northern portions of the Great Lakes. The region is one of moderate 
precipitation (60 to 115 cm.) and considerable temperature extremes, 
and the forest consists of a single association or associes dominated 

Fig. 99. — ' Lake-forest ' of southeastern Canada, dominated by White Pine (Pinus 
strobus) and Hemlock (Tsuga canadensis), in summer. The undergrowth is mixed 

and relatively sparse. 

by White Pine (Pinus strobus), Red or Norway Pine (P. resinosa), 
and Hemlock (Tsuga canadensis). Associated are various broad- 
leafed deciduous trees of various ecological affinities, making this 
forest all the more difficult to delimit ; indeed it is now often ques- 
tioned whether it ought to continue to be recognized as a type. 
For whether or not it is climax, it is in many respects transitional 
between the boreal coniferous and southern deciduous forests, from 
whose migrational buffetings and competition it has suffered. This 
is reflected in the undergrowth, which tends to be poorly developed 
owing to the dense canopy but includes many Pteridophytes, 


saprophytes, and under-shrubs. Fig. 99 shows an example of this 
type in summer, and Fig. 100 shows the selfsame area in winter. 

5. Besides the above we should mention some other Conifer- 
dominated types such as the montane and subalpine forests of 
western North America which, as their names imply, are due in 
part to local physiographic factors ; also the various ' Pine-barrens ' 

Fig. 100. — The same area of ' lake-forest ' as that shown in Fig. 99, but under 
winter conditions, with snow covering the ground. 

and other characteristic communities of eastern North America, 
which are due in part at least to fire or other disturbance, and 
consequently are of serai or subclimax nature. Somewhat com- 
parable but apparently still more often edaphically engendered types 
exist in various parts of Eurasia south of the boreal forest belt, or 
as outliers for example in the Mediterranean region. 

Warm -Temperate Rain Forests 

In warm-temperate as in subtropical regions where rainfall is 
plentiful and well-distributed through the year, evergreen forests 
are developed. The total precipitation is usually between 150 and 
300 cm. per annum and frosts are no more than occasional and slight. 
Towards the tropics these hygrophilous (i.e. moisture-loving) forests 
merge into the subtropical and finally the tropical types described 


in Chapter XIV, but in cooler regions they partake more of the 
characteristics of the deciduous summer forests described above. 
In spite of the relative abundance of climbers and epiphytes, even 
the human inhabitant of cool regions can scarcely consider many of 
these warm-temperate rain forests as anything like tropical, and so 
it seems desirable to treat them here and correspondingly reduce 
the width and complexity of our tropical, etc., belt. At best these 
temperate rain forests tend to be considerably less luxuriant as well 
as usually lower than the tropical ones, and to have fewer climbers, 
epiphytes, and other ' forest furnishings '. Nor are plank-buttresses 
(see Fig. 145) normally found in them. Moreover, they often show 
a fairly sharp distinction between winter and summer aspects, for 
example through admixture of deciduous trees. Although the 
co-dominants are often numerous and inclusive of Conifers, the 
local dominance is usually less mixed than in the tropics. The 
leaves tend to be smaller and more leathery, and the main canopy 
less dense, so that in different places Tree-ferns, smallish Palms, 
Bamboos, small trees, tall shrubs, etc., form a lower tier. Often 
the undergrowth is very dense and intertwined with herbaceous 
climbers, the ground and tree-trunks being covered with a mat of 
cryptogams and small herbs, making the whole quite difficult to 

Warm-temperate rain forests are developed sporadically in the 
southern portions of the United States bordering on the north shore 
of the Caribbean, and more extensively in southern Japan and 
adjacent Korea as well as westwards deep into China, in south- 
western South America, in the extreme south of Africa, and in New 
Zealand and some adjacent parts of Australia. Types approaching 
them are also found in uplands in the tropics, for example of southern 
Asia. In some places they merge into the sclerophyllous types that 
are commonly developed in the less summer-humid of the warm- 
temperate regions and will be considered under the next general 
heading. Many different faciations characterize different regions, 
particularly, and could be distinguished for example on the basis 
of different dominants. They may also be marked by characteristic 
associates, as in the case of those of Australasia with their Tree-ferns 
and those of the southeastern United States with their festoons of 
Spanish-moss (Tillandsia usneoides — see Fig. 101). Four examples 
from different quarters of the globe may be briefly described. 

1. The rain forest of southern Japan which, where undisturbed, 
is largely dominated by several species of lofty evergreen Oaks. 



Associated are other trees, including members of the Laurel and 
Magnolia families, and numerous shrubs forming a dense under- 
growth. Woody climbers (lianes) are plentiful, as are epiphytic 
Ferns and some Orchids. 

2. The temperate evergreen forest of the southeastern United 
States (but not southern Florida which is subtropical) where again 
more or less evergreen Oaks may predominate — especially Live 
Oak (Quercus virginiana) in the so-called ' hammocks '. Here the 
Evergreen Magnolia {Magnolia grandiflora) is often prominent. 

Fig. ioi. — Warm-temperate rain forest in southeastern United States. Live Oak 

{Quercus virginiana) and other trees are festooned with Spanish-moss {Tillandsia 

usneoides). Aquatic vegetation inhabits a sluggish stream in the foreground. 

There are a few lianes and temperate Palms in the rich shrubby 
undergrowth, and on the trees may be a fair range of herbaceous 
epiphytes among which the so-called Spanish-moss frequently 
dominates the landscape (Fig. ioi). However, the true broad-leafed 
forest is rather little represented because of edaphic and biotic and 
especially fire influences. These lead to subclimax areas occupied 
respectively by Bald-cypress (Taxodium distickum—see Fig. 102) or 
other summer-green swamps or, on dry sands, by evergreen Pine 
forest or savanna. Prominent Pines in this connection are the 
Loblolly (Pinus taeda), Longleaf (P. palustris), and Slash (P. 

3. The rain forest in New Zealand, which is almost entirely 


temperate in nature, in spite of the prevalence of large Tree-ferns. 
Stately Conifers such as the Kauri (Agathis australis) and various 
species of Podocarpus, and huge dicotyledonous trees with leathery 
leaves, are among the various and usually mixed dominants, with 
species of the smallish-leafed Southern Beech (Nothofagus) especially 
in the less luxuriant upland regions of the south. Although bright 

Fig. 102. — Bald-cypress (Taxodium distichum) swamp in southeastern United 

States. The dominant trees have ' breathing ' roots growing up into the air. 

Their branches are heavily festooned with Spanish-moss. In the foreground are 

floating-leaf and other early stages of the hydrosere, 


flowers are largely absent in this forest, the aspect is one of consider- 
able luxuriance owing to the profusion of shrubs, epiphytes, and 
climbers— as well as of Ferns and lower cryptogams. 

4. The temperate rain forest of southern Chile, which may be 
almost impenetrable owing to the density of the undergrowth. The 
dominant trees are again various and usually mixed, including some 
small-leafed evergreen Southern Beeches and other types, and a few 
Conifers. Bamboos of the genus Chusquea are often important in 
the dense undergrowth, and a considerable variety of climbers and 
epiphytes are again to be found. 


In warm-temperate regions having a rather hot and dry summer 
alternating with a cooler moist season, the dominant trees and shrubs 
tend to be evergreen and to have small and hard, thickish leathery 
leaves (sclerophylls). This is characteristic of most of the shores 
and hinterland of the Mediterranean, after which such climates are 
commonly named, though they occur also in the southwestern 
portions of Australia and South Africa, in central-southern and 
southeastern Australia, in the extreme southwest of the United 
States and adjacent Mexico, and in central Chile. Although thev 
sometimes abut on areas of warm-temperate rain forest, these 
sclerophyllous forest areas tend to show greater daily and seasonal 
temperature extremes, with snow and ice not infrequent around 
midwinter. Moreover the precipitation, besides falling irregularly, 
is usually much less plentiful than in rain-forest areas, typically 
ranging from 50 to 100 cm. (approximately 20 to 40 inches) yearly, 
while the quantity of moisture in the air varies greatly at different 
times. The plant communities tend to be rather drab through most 
of the year but attractively decked at flowering time. 

Although shaded uplands and areas of sufficient ground-moisture 
may bear more luxuriant mixed or deciduous forests, the typical 
dominants of sclerophyllous regions are lowish and gnarled, examples 
being rounded or flat-crowned evergreen Oaks, Olives (Olea spp.), 
or similar trees, and needle- or scale-leafed Conifers. They usually 
grow in more or less open, scattered formation, at least after dis- 
turbance by Man, and when destroyed are commonlv replaced bv 
a fairly dense scrub of mixed deciduous and evergreen bushes — 
including members of their own undergrowth. Even the associated 
herbs often have the form of shrub-like perennials, or store water 


in massive aerial tissues in the cases of Aloes, Agaves, and Cactus- 
like or other succulents. Particularly characteristic are a host of 
geophytes with underground food-stores in bulbs, tubers, etc., ready 
to develop with the rains. The examples mentioned above in five 
different continents may be considered briefly. 

1. The thin sclerophyllous woodlands of the Mediterranean and 
southern Black Sea regions. The chief dominants include evergreen 
Oaks such as the Cork Oak (Ouercus suber) and Holm Oak (Q. ilex), 
and various Pines such as the Aleppo Pine (Pinus halepensis) and 
Stone Pine (P. pined). However, most areas have been so disturbed 
by felling and grazing that only sparsely scattered, low and gnarled 
trees remain, the prevailing vegetation being a pale scrub on lime- 
stone terrain, known as ' garigue ', and a denser and taller one on 
siliceous soils, known as ' maquis '. This last is often 3 metres or 
so in height and in various forms and densities covers vast areas, 
being composed of a bewildering variety of shrubs including the 
subdominants of the original woodlands — such as Cistuses (Cistus 
spp.), Olive (Olea enropaea, extensively cultivated as a tree), Myrtle 
(Myrtus communis), Rosemary {Rosmarinus officinalis), Lavender 
(Lavandula latifolia), and tall Heaths (Erica spp.). Some Palms 
and large Cactus-like succulent Euphorbias may also occur. 
Epiphytes are generally absent and climbers few, but any open 
ground tends to support numerous bulbous or tuberous Mono- 
cotyledons, xerophilous Grasses, dicotyledonous herbs, and short- 
lived spring annuals in great variety. Fig. 103 shows a rocky area 
in open sclerophyllous woodland, with patches of subdominant 
mixed scrub of maquis and garigue sorts. East of the Mediter- 
ranean this type thins out with decreasing precipitation, although 
some semblance of it is still to be seen on the lower slopes of the 
mountains of northern Iraq, so invoking the interior of Asia. 

2. The Cape region of South Africa of which it has been written : 
' There seems to be no doubt that [it] once possessed luxuriant 
forests of the Mediterranean type, and that the same process of 
destruction which gave origin to the European maquis largely also 
transformed these forests into mere brushes ' (Hardy, The Geography 
of Plants, p. 246). Now there remains chiefly a wealth of sclerophyl- 
lous shrubs such as species of Protea and Leucadendron, with 
numerous tall Heaths and other bushy perennials belonging to very 
various families, and bulbous and tuberous subordinates. 

3. The sclerophyllous woodland and scrubby ' chaparral ' com- 
munities of western California and some adjacent regions. Here 

Fig. 103. — Rocky area with patches of mixed scrub of ' maquis ' type (' garigue ' 

where light-coloured), with herbs in open tracts and sparse sclerophyllous, etc., 

woodland in background, in the Mediterranean island of Corsica. 

FlG. 104. — Sierran chaparral climax, Santa Barbara, California. (By permission 
from Plant Ecology, by Weaver & Clements, copyright date 1938, McGraw-Hill 

Book Co.) 

35 6 



the dominant trees are often sparse and include Conifers and several 
species of evergreen Oaks, which in very dry situations are reduced 
to mere shrubs. The main mass of vegetation is commonly a thick 
and often tall scrub composed of representatives of very diverse 
families, with some associated succulents and numerous bulbous 
and tuberous herbs. Over some considerable areas trees are absent, 
this being the characteristic chaparral (Fig. 104), while, in others, 
only occasional isolated gnarled trees rise above the maquis-like 
scrub. Even where they form a canopy, the trees are typically only 
20-40 feet high. 

105. — Sclerophyllous forest in Australia, dominated by lofty Gum-trees 
{Eucalyptus spp.). 


4. Very similar conditions and vegetati6n-types exist near the 
coast of central Chile, where maquis-like scrub is widespread and 
gives way, inland, to slopes of sclerophyllous woodlands often 
dominated by trees strongly resembling evergreen Oaks. Their 
systematic allegiance is, however, usually quite distinct, as is that 
of the associated shrubs. In these woodlands, climbing plants as 
well as tuberous and bulbous ones may be common. 

5. The sclerophyllous woodlands and scrublands of southwestern, 
central-southern, and southeastern Australia are again in many 
cases reminiscent of those of the northern hemisphere, though their 
systematic allegiance is for the most part entirely different, usually 
involving other genera or even families. The most luxuriant expres- 
sion of the woodlands is found in the majestic forests of Gum-trees 
(Eucalyptus spp.), with an abundant undergrowth of hard-leafed 
shrubs often having beautiful flowers, and including Acacias, 
Mimosas, and Heath-like Epacridaceae. In many places Grasses 
cover the ground, the trees being scattered or letting through 
plentiful light owing to their leaves (or leaf-like members in the 
case of the Acacias) having a tendency to being orientated parallel 
to the rays of the sun. With less rainfall or more disturbance, 
the vegetation may be reduced to maquis-like ' mallee ' or other 
monotonous, tangled scrub most often 1-3 metres high. In areas 
of intermediate water conditions this may be interspersed with 
stunted Eucalypts, Casuarinas, and various other trees. Fig. 105 
shows such a sclerophyllous forest dominated by tall Gum-trees in 
a region having 75-100 cm. of rainfall annually. 

Heathlands and Grasslands 

Areas dominated by characteristic members of the Heath family 
(Ericaceae), or by narrow-leafed Heath-like shrubs, are common in 
temperate and adjacent lands, as well as in arctic and alpine regions 
beyond the limits of arborescent growth. Typically the com- 
munities are dense and of the order of 25 cm. in height. In warm 
lands they may merge into the scrubby types of sclerophyllous 
vegetation, which often include large Heaths, etc. But their most 
notable development is in cooler regions with a moist winter — 
particularly in western and north-central Europe, where the Common 
Heather or Ling (Calluna vulgaris) is often the chief constituent of 
the vegetation over considerable areas, particularly on acidic soils. 
Associated may be various Heaths (Erica spp.), Bilberries etc. 


(Vaccinium spp.), and Bearberries (Arctostaphylos spp.), with a 
variety of Grasses and sometimes some taller evergreen shrubs. 
Especially in the North, the heath-like Crowberry (Empetrum 
nigrum s.l.) is apt to be important and similarly gregarious, as are 
taller but still low and shrubby deciduous Birches and Willows. On 
the other hand in the South, the taller shrubs are often evergreen 
and characteristically have needle-like leaves (e.g. Juniper, Juniperus 
communis s.l.) or other photosynthesizing members (e.g. Gorse, Ulex 
spp.). The tendency to be gregarious and mycorrhizal is almost 
general in these plants : most of the chief dominants at least in the 
north being moreover evergreen, dwarfed, and richly branched 
chamaephytes, a characteristic dense and dark mat up to a ^-metre 
in height commonly results. 

Whereas most low-lying heathlands in temperate regions are 
subclimaxes of the disclimax type, being due to intensive grazing 
or recurrent fires that prevent trees from returning to their once- 
forested areas, in some coastal tracts especially of northwestern 
Europe these heathlands are evidently maintained through exposure 
to winds. In such instances they will be in more delicately balanced 
equilibrium with the environment and consequently more ' natural \ 
Much the same delicate balance and relative stability probably 
obtains in many upland areas, where damper peaty ' moors ' are 
especially common in cool regions — including the ' highmoors ' (cf. 
Fig. 168) developed on acidic soils inhabited by Bog-mosses 
(Sphagnum spp.), and the ' meadow-moors ' developed on circum- 
neutral (usually calcareous) soils. 

Still more important and widespread are grasslands, which indeed 
in their various forms constitute one of the main ' world ' types of 
vegetation. Whereas in tropical and subtropical regions grasslands 
typically take the form of savannas, with widely-spaced trees and/or 
tall shrubs as described in Chapter XIV, in temperate lands they are 
usually without trees or bushes except along watercourses. And 
although many grasslands are due to interference by Man or his 
domestic animals, many others, including extensive ones in temperate 
and allied areas, seem to be entirely ' natural '. Thus if they are 
due to grazing this is, or at all events originally was, apparently by 
wild animals. 

What seem to be climax grasslands develop in temperate regions 
chiefly in areas having an average yearly precipitation of between 
25 and 75 cm. (approximately 10 and 30 inches), or rather more in 
warm parts. These grasslands occur especially in the interiors of 


the great land-masses, the main examples in principally cool regions 
being the North American ' prairies ' and the Russian and adjacent 
' steppes ', with, in somewhat warmer regions, the South African 
1 veld ', South American ' pampas ', and southern Australian and 
New Zealand grasslands. These last three regional types frequently 
bear isolated or sometimes aggregated low trees or shrubs and accord- 
ingly constitute savannas ; indeed much of southern Australasia, 
particularly, is often indicated as savanna on world vegetation-maps. 
Nevertheless the real dominants are usually Grasses, so such types 
seem best classed here. In addition, meadows and other grasslands 
occur widely in temperate and allied regions as biotic plagioclimaxes. 

All grasslands have this in common, that they are dominated over 
at least most of their often vast area by Grasses—usually by various 
and often mixed perennial species which are narrow-leafed hemi- 
cryptophytes, many of them being gregarious and extremely hardy. 
Characteristically, the rooting is shallow and the underground parts 
form a matted turf which holds rainwater when and where it falls, 
checking penetration to deeper layers and accordingly aiding the 
grassland plants to prevail in their competition with trees in transi- 
tional areas where the rainfall is barely sufficient for arborescent 
growth. The turf and sod of old dead leaves, etc., may also help 
to check the successful establishment of tree seedlings. The winter 
in the so-called climatic grassland areas is often severe and dry, as 
may be the later summer; but recurrent spring and early summer 
rains that more than compensate for evaporation will favour the 
Grasses, which can then vegetate actively. 

The main types of grasslands occurring in temperate regions may 
now be briefly described. 

1. The prairies of central and western North America, ranging 
from well north in Canada southwards into Mexico. The dominant 
Grasses mostly form clumps (' bunch Grasses ') in the drier regions 
or more extensive sodded swards in the less arid ones, being inter- 
spersed with a large variety of subdominant forbs (herbs of other 
than grass habit). These often give the prairie a distinctive tone 
locally and fall into mixed societies flourishing for example at different 
times of the year. In general, however, various greens prevail in 
the early part of the growing-season, and yellows or greys later on, 
sometimes followed by darker orange or other autumn tints. Woodv 
plants are usually few and unimportant, except in depressions or 
as a response to overgrazing, which may also favour Cacti. Yet 
differences from spot to spot tend to be so marked and unstable 


that the whole has been claimed to constitute ' an open system ' in 
which repeated readjustment to disturbance overrules any tendency 
to equilibrium. In general correspondence with decreasing rainfall, 
the Grasses fall into three groups based upon stature and known 
respectively as ' tall ', ' mid ', and ' short ' Grasses. They range 
from the height of a Man, or taller, down to a few centimetres in the 
cases of the most drought- or pasturing-resistant types. Fig. 106, A, 
shows an area of true prairie in Nebraska, dominated principally 
by mid Grasses, but with shrubs and even trees in some damp 
depressions. Fig. 106, B, illustrates an area of short-grass prairie in 
Colorado, while Fig. 106, C, shows a similar area that has been badly 
overgrazed. In the prairies there is a long resting stage each year, 
which is due to low r temperatures in the North — the melting of winter 
snow affording plentiful water in early spring — and to low rainfall 
in the South and West. The general unity of this grassland climax 
is evidenced by some of the grass species occurring in nearly all 
of the component associations, and by the large number of grass 
genera — such as Agropyron, Bonteloua, Elymus, Poa, Sporobolus, and 
Stipa — that afford dominants more or less throughout its great 
range. Nevertheless, differing local conditions lead to preclimax 
and postclimax communities and different treatments to biotic 
plagioclimaxes, while near the forest border is an ecotone with trees 
which elsewhere persist chiefly along watercourses. In such zones 
of tension, each life-form takes advantage of the slightest variation 
of edaphic or biotic impress which may be in its favour. 

2. The steppes of the U.S.S.R. and adjacent lands, covering vast 
areas south of the northern forests and north of the central deserts, 
etc., and ranging from eastern Europe to eastern Asia, with outposts 
farther west and south. These steppes are similar to the North 
American prairies in all essential respects, though they differ con- 
siderably in the component genera and much more in the species. 
The Grasses tend to be highly xerophilous and to form dense tufts 
or cushions composed of the remnants of previous years' growth 
seated on stools of superficial but much-branched fibrous roots, 
while the associated forbs are mostly hardy perennials or bulbous 
geophytes. These associates and any annuals or woody plants are 
mostly small, few exceeding half-a-metre in height. The growth 
of trees is almost everywhere prevented by the scarcity of water 
and by the extremely severe winters, when strong and dry winds 
blow over the frost-bound soil on which snow may afford protection 
only for ground-vegetation. Between the northern forests and open 

FlG. 106. — The North American Prairie: mid- and short-grass communities. 
A, Nebraska prairie area dominated principally by ' mid ' Grasses, with shrubs 
and trees in depressions. (By permission from Plant Ecology, by Weaver & 
Clements, copyright date 1938, McGraw-Hill Book Co.) B, short-grass com- 
munity in Colorado. (Phot. U.S. Forest Service.) C, area similar to (B), but 
overgrazed and in poor condition, with bad weeds, including Cacti. (Phot. U.S. 

Forest Service.) 



steppes there is often a broad ecotone, rather incongruously known 
as ' forest steppe ', in which the two formations occur intermixed 
in patches. 

3. The Argentinian and other South American pampas and 
steppes, again developed chiefly on flat or gently undulating plains, 
and with most of the general characters and range of variation of 
the above. Quite frequently some low trees or tall shrubs give 
these grasslands the aspect of savannas. 

4. The grasslands and savannas of South Africa and southern 
Australasia — the grasslands being again closely comparable with the 
northern steppes and prairies, whereas the savannas have also 
scattered trees and or bushes. But although the dominant Grasses 
tend to be of familiar forms and even genera, the woody plants are 
very different in different regions, often including unique types 
[e.g. in Australia). 

5. The subclimax meadows and other grasslands of various 
temperate regions that except for biotic disturbance would support 
vegetation of higher life-form. These probably include considerable 
areas of present-day steppes and prairies especially towards their 
forested margins, but are more notably represented by the verdant 
meadows that form such a prominent feature of the dairy and other 
pastured lands, for example, on either side of the North Atlantic. 
Such meadows are typically due to clearance of the forests and 
more or less long-continued mowing and/or pasturing by domestic 
animals and wild herbivores such as Rabbits and Hares. This 
favours the Grasses and other hemicryptophytes that dominate such 
areas, so that they tend to form a continuous sward with closely- 
compacted turf which represents a biotic plagioclimax in that it is 
deflected from the normal succession. However, with removal or 
reduction of normal grazing, woody plants soon enter the system 
and the subsere proceeds towards the climax. Much the same 
Grasses and forbs with buds hidden in the surface layer of soil are 
protected against fire, which often helps to maintain grasslands, as 
does the tendency of the turf to retain water. In meadows, the 
dominant Grasses are usually broader-leafed and less xerophilous 
than in climatic grasslands, although a wide range of often similar 
types and even genera are involved. Meadows, moreover, tend to 
support a wider variety and often greater admixture of forbs. Most 
of the meadow inhabitants are more or less hygrophilous, lacking 
marked protective devices reducing transpiration, and, although 
their flowering-axes die down, many remain uninterruptedly green in 


winter. The dominant Grasses are perennial, with axes on the 
average about half-a-metre high, and typically involve such familiar 
genera as Poa, Festuca, Lolium, Agrostis, Alopecurus, Phleum, and 
many others. The associated forbs are also mostly perennial, or, 
less frequently, biennial, and are often rosette-chamaephytes. Like 
the Grasses, they may be coarse and 1-2 metres high in favourable 
circumstances especially in damp situations, but rise scarcely at all 
above the surface of the soil when intensely grazed ; such close 
cropping cannot, however, be long withstood by most species. 
Annuals occur chiefly in open patches that are due to overgrazing 
or drought, while Mosses may form a weak ground-layer especially 
in the damper situations. Woody plants, like geophytes, if present 
are usually little in evidence — except where the biotic disturbance 
is light and there is a tendency towards heath development or 

Semi-Deserts and Deserts 

Although what amount practically to desert conditions are com- 
monly developed very locally on dry rock and sand or other porous 
surfaces, these usually show evidence of at least incipient succes- 
sional change. On the other hand, extensive desert areas tend to 
be among the most stable vegetationally, owing to the general lack 
of water for more growth than that of the plants already present. 
Intermediate in water-relations between true deserts and the drier 
among the areas whose vegetation has been described earlier in this 
chapter, and usually intermediate also in geographic position, are 
various semi- or near-desert areas such as the Sage-brush ones of 
western North America. These are characterized bv the Sage-brush 
(Artemisia tridentata) and other low and often greyish shrubs having 
herbaceous branches, that dominate a climax in regions having an 
annual precipitation of 12-25 cm - (approximately 5-10 inches). 
Such shrubs often belong to predominantly herbaceous families and 
also thrive in areas of greater rainfall when the competition from 
Grasses is eliminated. The scrub is low and often broken or of 
more or less widely-spaced single bushes, the depauperated tvpes 
merging into the still more xerophytic ' desert-scrub ' developed 
where the rainfall is limited to 7-12 cm. annually. In this latter 
instance the dominants are bushy shrubs, particularly of Creosote- 
bush (Larrea tridentata), usually \-z metres high and spaced on 
the average 5—1 5 metres apart through a widely-spreading root 


In the often comparable ' desert steppes ' of the U.S.S.R., which 
constitute the transition between the more or less continuously 
vegetated steppes and the deserts of predominantly bare spaces, 
other Wormwoods {Artemisia spp.) of Sage-brush type tend to be 
the most characteristic plants. Frequently they are halophytic, the 
soils being salinized and supporting also such characteristic under- 
shrubs as Salt-bush (Atriplex canum). Often prominent in addition 
are sod-forming Grasses such as wiry Fescues (Festuca spp.) and 
Feather-grasses (Stipa spp.), and dicotyledonous as well as mono- 
cotyledonous annuals which develop rapidly with spring rains but 
die down with the advent of hot weather, so leading an ephemeral 
life. Examples of this last phenomenon are well seen in desert areas 
of Iraq— cf. Fig. 156. 

Some Lichens may grow or lie on the surface of the earth, as may 
Blue-green Algae including colonies of Nostoc, but arborescent 
growth, for example of Willows and Poplars, is usually limited to 
occasional damp depressions or the vicinity of watercourses. Such, 
for example, are the oases of the Gobi Desert. Rather similar 
conditions and attendant vegetation, with a dry and generally warm 
climate subject to sharp contrasts of heat and cold, and one or two 
brief vegetative seasons each of 1-3 months (in spring and/or 
autumn) characterized by intermittent rainfall and favourable 
temperatures, are widespread in temperate as well as in warmer 
countries. Thus they occur on the plateaux of Asia Minor, in belts 
around the deserts of Central Asia and Australia, in southern South 
America and South Africa, and flanking the Sahara and Arabian 
Deserts. Some of these areas support extensive thorny or other 
scrub and are apt to be so designated on world vegetation maps. 

These semi-desert regions of both the Old and the New Worlds 
are typically made up of vast, flat or undulating expanses of bare 
soil supporting hoary or dull-grey, often sticky and scented, ' heathy ' 
or sage-like shrubs J-2 metres high. Such shrubs may form a 
continuous but thin or else broken brush, or be ' scattered ' over 
otherwise naked tracts, though in particularly arid areas the spacing 
is often more even and apparently due to root-competition. Al- 
though there are no green grassy swards there may be bunches of 
wiry Grasses, and although there are normally no trees there may be 
tallish Cacti or giant Euphorbias or other large succulents, while 
saline lagoons, dry most of the year, may be surrounded by fleshy 
or bushy halophytes. The usually erect or ascending, switchy 
dominant shrubs often have thickish woody stumps and deep roots. 


Their leaves are commonly narrow, leathery and inrolled, less than 
3 cm. long and often densely covered with hairs which give them a 
greyish-white tint. Outside the rainy season, dependence is largely 
on underground water. Except for those which are deciduous, the 
dominants frequently show little contrast in appearance at different 
times of the year — especially as few display at all large and bright 
flowers. Matters are otherwise with the often plentiful associated 
ephemerals which burst forth when rain allows, and form quite a 
display with such succulents as Cacti, Mesembryanthemums, Aloes, 
and Agaves, which contain considerable stores of water. 

With still more precarious precipitation or other water supply, 
deserts usually result. But as these usually belong to tropical or 
subtropical regions, or at least extend well into them, they are treated 
chiefly in Chapter XIV. Exceptions are afforded by the Trans- 
caspian desert which is an open plain and the Gobi Desert which 
is a plateau, both lying in temperate parts of Asia. These are 
mainly areas of prevailing drought and extreme temperature con- 
ditions, and consequently are very little vegetated, such plant life 
as exists being mainly the result of depauperation of the so-called 
desert steppes described above. Thus dunes may be partly fixed 
by low and open, shrubby growth, but siliceous and clayey soils, 
often containing loess, are apt to be virtually barren, as are gravelly 
and talus-strewn areas over considerable tracts of country. Nor do 
the frequent saline or ' alkali ' and gypsiferous areas afford much 
relief to the abiding monotony. Nevertheless one seldom finds at 
all extensive tracts even in the Gobi that are entirely devoid of some 
kind of vegetation, and very commonly the transition to steppe or 
at least desert-steppe is marked by a scattering of poor Grasses 
ranging from some 25 cm. high in exposed situations to a metre in 
height in depressions where water collects in the rainv season. 
Besides Grasses and some Sedges, xeromorphic members of the 
Daisy (Compositae), Goosefoot (Chenopodiaceae), and Tamarisk 
(Tamaricaceae) families tend to be prominent in these temperate 
deserts, as do geophytic monocotyledons such as Tidipa uniflora, Iris 
sisyrinchium, and species of Gagea. Desert-like ' bad-lands ', often 
characterized by lowly Cacti or brush-like shrubs, also occur more 
locally in parts of temperate North America and southern South 

For examples of desert areas in warm-temperate regions we may 
go to central Iraq. Fig. 156, A, shows a quadrat in a typically 
gravelly area in which one small perennial tuft is visible but there 

<• ^jSftK 

Fig. 107. — Salt desert and salt-marsh of a warm-temperate region. A, an area 
of salt desert near Shithatha, in southern central Iraq. Clumps of the co-domin- 
ants are up to 2 metres wide and 60 cm. high. Cracking of the surface encrustation 
is noticeable in slight depressions. The dots at the foot of the distant scarp are 
grazing Camels. B, saline marshes outside the oasis of Shithatha, Iraq. The 
vegetation is mostly closed and sedgy-grassy, with some Tamarisks (Tamarix spp.). 



are numerous tiny seedlings springing up after heavy rain. Fig. 
156, B, shows a close-up of a near-by less gravelly area when the 
ephemerals have developed. Fig. 107, A, shows an area of salt 
desert near the oasis of Shithatha, in southern central Iraq, where 
higher vegetation occupied from one-eighth to one-half of the sur- 
face, bushy Chenopodiaceae affording the main dominants. Soluble 
salts totalled 7-5 per cent, of air-dry soil at the surface where tested 
but decreased markedly below ; the pH was 8-3 at the surface and 
scarcely varied from this below. Some widish and flat, white- 
encrusted areas tended to be barren, though in places where water 
could collect in the rainy season and Blue-green Algae grew, loose 
mud ' medallions ' covered the surface. 


Whereas various salts, such as nitrates, sulphates, and phosphates 
of potassium, calcium, magnesium and iron are essential, at least 
in the small concentrations that are usually present in soils, for the 
normal development of land vegetation, excessive salinity (e.g. of 
more than 0-5 per cent.) is harmful to the growth of most plants. 
Such high salinity, particularly due to sodium chloride, is most 
commonly found around sea-shores, and is the most constant factor 
leading to the replacement there of normal land-vegetation by plants 
which habitually grow in very salty soils (halophytes) or at least 
can grow in such soils (facultative halophytes). Thus in maritime 
salt-marshes, which are primarily determined by periodic immersion 
in salt water and mostly lie between the levels reached by the higher 
neap and ordinary spring tides, the components of the characteristic 
vegetation are almost all peculiar to the habitat. And whereas, if 
the tide is effectively excluded and there is adequate drainage, the 
salt in time is washed out of the soil and non-halophilous species 
colonize the area, ' There is no good evidence that salt marsh can 
develop by the mere accumulation of silt or humus, without human 
assistance, into a non-maritime vegetation ' (Tansley, Introduction 
to Plant Ecology, p. 78). Consequently the salt-marshes of tem- 
perate and allied regions seem best considered here among subclimax 
or local climax types. 

Salt-marshes are chiefly developed on mud-flats about sheltered 
tidal estuaries. In them such Green Algae as Rhizoclonium and 
Enter omorpha, or such halophilous vascular plants as succulent or 
shrubby Glassworts or Saltworts (Salicornia spp.), are the normal 


pioneers, followed typically by Alkali-grasses {Puccinellia spp.) or 
other halophytic Grasses. Any such growth tends to impede the 
flow of water and increase the deposition of silt, helping to raise 
the level of the bed. Higher up, on the ' flats ' covered by most of 
the spring tides, a mixed vegetation is usually developed. This is 
the general salt-marsh community which, for example on both sides 
of the North Atlantic, characteristically includes such types as Sea 
Plantain (Plantago maritima s.l.), Sea Arrow-grass (Triglochin 
maritima), Sea-blite (Suaeda maritima), and Sea-pink (Armeria 
maritima s.l.), with shrubby Chenopodiaceae especially in warm- 
temperate regions. Various Brown and other Algae, often of peculiar 
habit and squalid mien, are commonly associated. Still higher up, 
where the saline flats are covered only by the higher spring tides, 
a more or less close turf usually develops with the help of grazing, 
composed of halophilous Grasses with associated forbs. At the 
uppermost levels reached only by the very highest spring tides or 
storm-waves, less markedly halophilous types such as Red Fescue 
(Festuca rubra s.l., a facultative halophyte) tends to predominate, 
with often an admixture of ordinary land species that can tolerate 
small amounts of sea-salt in the soil. At the higher levels in river 
estuaries the water is alternately fresh and brackish with the ebb and 
flow of the tide, so inhabiting plants must be physiologically special- 
ized to withstand rather sudden and extreme changes in the osmotic 
value of the inundating water. 

Whereas such geodynamic factors as tidal action may cause the 
maintenance of a condition of equilibrium between accretion and 
erosion, and thus of a lasting type of vegetation in maritime salt- 
marshes, it seems probable that in some instances at least there is 
gradual continuation of accumulation and advance towards normal 
land vegetation — even without human interference — as in the case 
of sand-dunes and shingle beaches, and in spite of what was quoted 
above. But at all events the types just described are commonly 
both long-lasting and distinctive, and consequently had to be 
treated as special entities. 

In many arid regions, for example of interior Asia and western 
North America, there occur salt lakes or smaller basins or seasonally 
dry ' alkali pans ' exhibiting a similar range of conditions of varying 
salinity which is, however, often more extreme than on sea coasts. 
Inland, these saline areas result from excessive and prolonged 
evaporation, changes in level and salinity in individual cases being 
seasonal instead of tidal, though they may vary rapidly with the 


weather. Still, the sequence of zones is often remarkably definite, 
and in a general way reflects not only the prevailing degree of 
salinity but also the poor aeration accompanying the increasing soil- 
moisture. Here again there is no certainty that succession will ever 
proceed to any proper climatic climax in the absence of biotic 
disturbance, especially as the tendency is for the salinity to increase 
rather than decrease with time. Consequently it seems best to 
consider the marginal vegetation as forming a zoned series of sub- 
climaxes so far as the general region is concerned (or edaphicallv 
limited climaxes of their own immediate areas). In any case it 
appears likely that the local allogenic conditions will have more 
effect upon the future of these inland saline areas than will anv 
autogenic serai tendencies. Fig. 107, B, shows an area of saline 
marsh outside the oasis of Shithatha, in southern central Iraq. The 
vegetation is mostly closed and sedgy-grassy, with some shrubby 
Tamarisks, being dominated by such types as Scirpus maritimns, 
Cyperus distachyos, Aeluropus littoralis, and Tamarix pentandra. The 
pH was 8-6 where tested and the area was said to remain damp 
throughout the year, having some shallow pools of open water at 
least in spring. Open mud ' polygons ' were bound by mixed 
Oscillatoria, Lyngbya, and other primitive Algae. 

Some hitherto productive cultivated areas in arid regions requiring 
irrigation have become too saline for the growth of crops — owing 
to prolonged evaporation leaving the salts of the ' raw ' irrigation 
water behind when no drainage was provided. In other cases the 
crops are limited to facultative halophytes such as Date Palms 
(Phoenix dactylifera) or Barley. Much of central and southern Iraq 
is of this nature, and as Man has not yet become proficient at 
remedying such soil salinity on a wide scale, the increasing amount 
of irrigation is worsening the situation. 

Seral Communities 

Besides subclimax communities, various other deflected or 
1 straight ' successional ones have been described or implied above, 
while many others are of such limited occurrence or importance 
(as in the cases of salt-sprayed coastal areas or the ' seniles ' occurring 
for example on fallen logs) that we can scarcely even mention them 
in this brief outline of the main vegetational types of temperate 
and allied lands. However, to complete our picture there remain 
a number of (mostly seral) communities that should be cursorily 


elucidated, some of these being sufficiently widespread to demand 
more attention. 

Notable are the characteristic types afforded by the sand-dune 
and shingle-beach successions which occur chiefly (but in the former 
case by no means entirely) near sea coasts. Thus sandy sea-shores 
that are wetted by the highest tides are commonly inhabited in 
sufficiently sheltered situations by a characteristic open community 
of halophytic shore-plants such as Sea-rockets (Cakile spp.), Sea- 
purslane (Arenaria (Honckenya) peploides agg.), and species of Orache 
(Atriplex). These tend to arrest any dry sand that is blown on to 
them, and, consequently, to form the basis of hillocks through which 
the shoots grow, the effect being cumulative in that the deeper 
the sand comes to be piled up, the higher the plants will grow, 
and vice versa. More important in this connection are the major 
dune-formers — coarse Grasses such as, particularly, Marram Grass 
(Ammophila arenaria) and Lyme-grass (Elymus arenarius s.l.) in north- 
temperate regions. These have similar powers of colonization and 
sand-accumulation but grow chiefly farther back from the sea and 
more extensively, binding the sand by the ramifications of their 
rhizomes and roots. This is illustrated in Fig. 93, which shows 
Marram Grass doing the binding on the coast of Maine. Meanwhile 
the motion of the surface particles is checked by the tall and usually 
tufted aerial parts, and other species, which cannot colonize moving 
sand, are enabled to establish themselves between the axes of the 
main pioneers. These secondary colonists typically include Lichens, 
Mosses, and less coarse Grasses, which finally bind the surface and 
consolidate the community. Maritime shrubs and in time climax 
forest typically follow if the land is not severely pastured or turned 
into golf links (for which old, grass-covered dunes are the traditional 
sites). Meanwhile there will have been produced some characteristic, 
though serai, vegetational types whose non-halophytic counterparts 
may occur far inland, as for example around the Great Lakes of 
North America. 

Shingle-beach vegetation has considerable affinity with that of 
dunes, especially when sand is admixed and the habitats are thereby 
rendered similar. Where sand is lacking, various Lichens may 
colonize the surface of the stones inland of the usual ' storm-crest ', 
and over them may extend clumps of such maritime flowering plants 
as the Sea-pea (Lathyrus maritimus agg.) and halophytic Grasses. 
These often persist for a long time, the plants being rooted in the 
crevices between the pebbles. Ultimately the crevices become filled 


with sand and plant remains, and gradually the vegetation extends 
to form a complete covering of the shingle, so that on old shingle 
beaches away from the direct influence of the sea, inland types of 
vegetation commonly develop, and in time scrub or even forest. 

Except that, usually, only inland plants are involved and the early 
stages tend to form long downward stabilized strips separated by 
dynamic tracts, much the same sequence takes place on many talus 
screes and heaps of detritus. These, when reasonably stable and 
including a fair amount of comminuted ' soil ', can, with the help 
of their surface crevices, run the whole gamut of the xerosere with 
relative speed and ease right up to a forest climax. Otherwise the 
lithosere is usually very slow in progressing, especially in its early 
stages. Although these stages may be represented by highly 
characteristic cryptogamic or herbaceous communities, these norm- 
ally occupy only very limited areas of rock, such as occur naturally 
in temperate regions chiefly on cliffs and rocky outcrops. 

The hydrosere also affords characteristic serai communities such 
as floating-leaf, reed-swamp, sedge-meadow, and damp scrub ones 
which were outlined in the last chapter and may severally cover 
considerable areas. Although they vary widely in different instances 
and especially in different regions within the temperate and allied 
belts, these communities are mostly too familiar and obvious to 
require description in any detail here. Dominants on both sides 
of the North Atlantic include Water-lilies (Nymphaea spp.) in the 
floating-leaf stage, Cattails or Reedmaces (Typha spp.) in reed- 
swamp, various Sedges (Car ex spp.) in the sedge-meadow, and 
Alders (Alnus spp.) in the damp scrub. Examples of most of these 
features are described and illustrated elsewhere in this chapter or 
the immediately preceding one (stages of hydroseres being shown 
especially in Figs. 89, 90, 91, and 102). 

Besides the various natural or semi-natural scrublands mentioned 
above, there are the often characteristic scrubs and thickets belonging 
to subseres that follow cutting or burning of forests in temperate 
and allied regions. These unstable types may occupy considerable 
areas, though usually not without more or less recent human dis- 
turbance or intervention. Even the later and taller stages of these 
secondary growths are frequently dominated by plants — such as 
' weedy ' Birches or Poplars — that do not occur in the climax 

There remain to be mentioned among tvpes naturally occupying 
substantial areas in temperate and allied regions the various marshes, 


fens, bogs, mires, moors, and waterside zones that commonly 
represent part of some hvdrosere or other. As such, many have 
been covered above, at least by general implication ; in other 
instances they seem to form subclimaxes, for example of plagioclimax 
nature where a continuing ' master factor ' is involved, and, at least 
when covering considerable areas, appear needful of some treatment. 
Outstanding are the marshlands, fenlands, and boglands which may 
be distinguished respectively according to whether the soil is formed 
mainly of silt, of peat containing considerable quantities of lime 
or other bases, or of peat very poor in such bases. With marshlands 
of one sort or another we are already fairly familiar, while further 
examples are mentioned in the following chapters. 

Fenland occupies the alluvial borders of rivers and streams as 
well as parts of old estuaries and the borders of certain lakes — 
especially of those into which streams bring ' hard ' water rich in 
lime. Its most characteristic manifestations represent stages in the 
hvdrosere — particularly reed-swamp, sedge-meadow, and, if the 
latter is not regularly cut or pastured, damp scrub or woodland 
' carr ' dominated by Alders or sometimes Birches, or constituting 
meadow-moors in exposed situations. The vegetation in these 
cases is largely calcicolous, whereas in the general run of alluvial 
marshlands it is less exacting. 

Bogs or ' mosses ', on the other hand, develop chiefly where the 
water is very poor in calcium and other basic salts, and support 
entirely different communities — for example around the shores of 
lakes and tarns in areas where the rock is deficient in basic mineral 
salts and the water is consequently ' soft '. Typically the growth 
consists largely of Bog-mosses {Sphagnum spp.) supporting a number 
of characteristic higher plants such as Cotton-grasses (Eriophorum 
spp.) and Sedges (Carex spp.), and often ' insectivorous ' associates 
such as Butterworts (Pinguicula spp.), Sundews (Drosera spp.), and 
Bladderworts (Utricularia spp.). In areas of cool and wet climate 
where the drainage is poor, such communities may cover considerable 
flattish tracts and be known as ' blanket bogs \ Owing to the Bog- 
mosses' remarkable power of holding water in their sponge-like 
cushions (see p. 51), these last can grow in height and enormously 
in width, extending over marshes and even fens or open water, 
depositing peat and raising the surface on which they grow. Such 
1 raised bogs ' tend to be strongly acidic in reaction and reddish- 
brown in colour. 

Although bogs are commonly colonized by Heaths or even larger 


woody plants, the Mosses seem to control 'matters as long as they 
remain fully active. However, with draining or natural drying out 
of the surface, the succession usually proceeds to the local forest 
or other climax — such as highmoors or drier heathlands in exposed 
situations. Lowland- or meadow-moors are usually hydroseral 
communities occurring on circumneutral peat accumulations result- 
ing from the filling up of lakes and preceding the timbered stages 
of succession. Here the upper layers above the level of ground- 
water are often acidic in reaction, and either support heathland 
communities dominated by Heaths in drier parts, or bog-like ones 
dominated by Cotton-grasses or Moor-grass (Molinia coerulea) in 
damper parts. Where acid conditions are developed by such humus 
accumulation above basic rocks, ' flushes ' from springs or superficial 
drainage may continually bring down fresh supplies of basic salts 
and lead to ' spring flush ' communities very locally. Especially in 
regions of marked topographical and geological variation is it common 
to find whole series of moorland, dry heathland, pastured grassland, 
and scrub or woodland communities existing side by side within a 
relatively small area. 

Besides the more extensive hinterland communities already men- 
tioned, those of sea and other cliffs, and of river and other banks, 
may be distinctive, though usually they are very limited in area and 
too variable for detailed consideration here. Notable, however, is 
the occurrence of cushion and other alpine plants along the dry 
margins or beds of watercourses in the lowlands of mountainous 
temperate countries on both sides of the Equator. Presumably their 
disseminules are washed down from the uplands and their growth 
is favoured by the ' open ' conditions and general lack of competition 
by ranker lowland types, which would prevent their ecesis or rapidly 
succeed them in most other habitats. 

Finally there are the various weed and allied communities that 
follow anthropic disturbance ; for such activities as cultivation or 
forest-cutting and their aftermaths are so widespread that in tem- 
perate regions there is relatively little truly natural vegetation left. 
Indeed of the more favourable areas capable of supporting climax 
forest there can scarcely remain any that are wholly undisturbed 
by Man or his domestic animals, though there are many which we 
think of (and study) as practically natural. Even crops may be 
considered to comprise communities of a sort, however artificial and 
temporary they may be. Representative crops and some weeds 
were dealt with earlier in this work ; especially do weeds form many 


and various if rather ephemeral communities, examples of which 
are all too familiar to every farmer, gardener, and estate-owner. 
Moreover, owing to almost universal introduction by Man and to 
the similar ' openness ' of the habitats involved, these colonies of 
weeds tend to be remarkably alike in similar situations throughout 
the temperate and allied regions, often involving the selfsame 
species in both Old and New Worlds, and in both Northern and 
Southern Hemispheres. But in spite of the common luxuriance of 
such weed communities, the full abandonment of agricultural or 
other waste areas usually allows succession to proceed so rapidly 
towards the local climax that within a very few years the weeds are 
liable to have disappeared entirely. These areas being usually in 
tracts that had been cleared of former forests, the weeds are com- 
monly superseded in the first few years by scrub or such weedy 
trees as Birches or Poplars, before the return of anything like the 
climax forest. 

Some Physiographic Effects 

Many physiographic effects, such as differences in exposure and 
water-conditions due to ridges and depressions, have already been 
dealt with, and the vegetation of uplands above the limit of arbor- 
escent growth is treated in the next chapter. Outstanding, however, 
are differences due to aspect and particularly altitude below the 
tree-line, which can lead to marked differences in conditions and 
local vegetation. This we have already mentioned and illustrated in 
Figs. 83, A, and 83, B, but must consider further here. Such differ- 
ences are largely due to differences in local climate as explained in 
Chapter X. The communities involved usually lie within the orbit 
of those described elsewhere and so need not be treated in any 
detail, though a few examples of the effects of (1) aspect and (2) 
altitude in temperate and allied regions may be given with advantage. 

The most widespread and commonly obvious aspect effect is that 
due to orientation with regard to the sun's rays. For example, in 
the Mediterranean region some ridges lying east and west may bear 
almost entirely different vegetation on their north- and south-facing 
slopes, in extreme instances having not a single ecologicallyimportant 
species in common. Thus the south-facing slopes, exposed to the 
full glare of the midday sun, tend to be occupied by the highly 
xerophilous maquis or garigue of more or less sparse shrubs and 
herbs. The ower parts of any steep north-facing slope, however, 


will be protected from at least the strongest insolation and may bear 
at the same altitude deciduous forest with hygrophilous ground- 
vegetation. On the other hand in moist northern regions the most 
luxuriant vegetation may be developed on the southern and western 
slopes which receive the greatest benefit from the sun. The tendency 
is of course reversed in the southern hemisphere. However, north- 
and-south-running mountain ranges may show instead another 
aspect effect, namely, marked differences in rainfall on their two 
sides. Thus of New Zealand the eastern side, sheltered from the 
prevailing westerly winds, has in places no more than one-tenth of 
the rainfall of the forested western side, and supports only poor 
tussocky grassland over considerable areas. 

The general tendency towards cooler and damper conditions as 
we ascend mountains usually leads to marked attendant changes in 
the vegetation. These may run the gamut from arid plains or 
lowland forests, dealt with above, all the way to high-alpine regions 
of perpetual snow. With the tundra and other communities lying 
above (as in latitudes beyond) the tree-limit, we shall be concerned 
in the next chapter. Those communities developed between such 
extremes and the general run of lowlands in temperate regions 
usually involve faciations or extensions of one or other of the forest 
types already described. However, in some instances, as most 
notably the desert uplands of parts of temperate Asia, there may be 
instead almost plantless wastes or expanses of moving sands — with 
salt tracts of various extent, and only occasional oases supporting 
deciduous trees such as Poplars. 

A good example of the usual forested sequence in mountainous 
districts is seen in western North America, where, above a basal 
zone of ' improved ' plains vegetation, the ' montane forest ' extends 
from the arid foothills upwards into the mountains through an 
altitudinal range of often some 2,000 metres. The main dominants 
are Ponderosa Pine (Pinus ponderosa), White Fir {Abies concolor), 
and Douglas Fir (Pseudotsuga taxifolia), though many others occur, 
the closest relationship being with the Pacific ' coast forest \ 
Extending above, through an altitudinal belt of commonly some 
1,000 metres, comes the ' subalpine forest ', which is related 
primarily to the boreal forest but also to the coast and montane 
forests, the main dominants being species of Picea and Abies — 
particularly P. engelmannii (Engelmann Spruce) and A. lasiocarpa 
(Subalpine Fir), with often some Lodgepole Pine [Pinus contorta 
var. latifolia) and allied species. While the multiplicity of dominants 




makes for varied groupings, there is a tendency towards pure con- 
sociations near the timber-line ; at its upper levels the forest also 
becomes less luxuriant and the canopy lower, until it passes into 
' elfin wood ' and ultimately ' Krummholz ' of stunted and twisted 
trees (see Fig. 108) about where the alpine tundra begins. How- 
ever, mountain vegetation has no uniform pattern but varies from 
range to range. Thus in many mountainous regions of the temperate 

Fig. 108. — Pine ' Krummholz ' at timber-line in the Rocky Mountains. (Phot. 

W. S. Cooper.) 

belt, as for example in central Europe and the White Mountains 
of New England, there is a tendency for the lower zone of montane 
forest, like the basal tracts, to be dominated by broad-leafed 
deciduous trees, and only the higher (subalpine) levels to be pre- 
dominantly coniferous. Often a fairly wide belt of mixed deciduous 
and evergreen forest intervenes ; and although there is a general 
tendency for corresponding zones to decrease in altitude towards the 
poles, there may be wide variation according to local conditions 
even on the selfsame line of latitude. 

Further Consideration 

Although this chapter represents in outline the results of considerable 
reading and personal experience, often in obscure periodicals or remote 


places, further details about most of the subjects treated may with 
advantage be obtained from the standard general works dealing with the 
vegetation of the land-masses of the world. These include : 

A. F. W. Schimper. Plant-geography upon a Physiological Basis, transl. 

and revised edition (Clarendon Press, Oxford, pp. xxx -f 839 and 4 

additional maps, 1903). See also the ' third ' German edition revised 

by F. C. von Faber and cited on p. 23. 
M. E. Hardy. A Junior Plant Geography (Clarendon Press, Oxford, 

pp. 1-192, 1913). 
M. E. Hardy. The Geography of Plants (Clarendon Press, Oxford, pp. 

xii -f 327, 1920). 
D. H. Campbell. An Outline of Plant Geography (Macmillan, London 

(and New York), pp. ix -f 392, 1926). 
A. G. Tansley & T. F. Chipp. Aims and Methods in the Study of 

Vegetation (Crown Agents for the Colonies, London, pp. xvi + 383, 

M. I. Newbigin. Plant and Animal Geography (Methuen, London, pp. 

xv + 298, 1936). 

Concerning pertinent ecological principles, and for some local examples, 
see the works of Tansley, Leach, Weaver & Clements, and Oosting, cited 
at the end of Chapter X. 

Treatment of the vegetation of different regions is extremely ' patchy ', 
even in the most populous of the temperate and allied lands as they were 
considered in the present chapter. Thus whereas on many areas there 
is an extensive literature, including numerous accounts prepared by 
trained and accomplished observers, on others there is very little. The 
total, however, is vast but scattered. Particularly have many valuable 
and often well-illustrated accounts appeared in the British Journal of 
Ecology, which has been published continuously since 191 3. Others have 
appeared from time to time in the American journal Ecology, which has 
been published regularly since 1920, in the companion Ecological Mono- 
graphs, of which a volume has appeared each year since its institution in 
1 93 1, and in the German Vegetationsbilder, of which 26 volumes were 
published during 1904-44. Further accounts are appearing in the more 
recently founded international journal Vegetatio (Junk, Den Haag, 1948-). 
Many of the pertinent papers are cited in S. F. Blake & A. C. Atwood's 
Geographical Guide to Floras of the World (see p. 214). 

A model of the kind of work which is desirable for each and every 
region is Sir A. G. Tansley's monumental The British Isles and Their 
Vegetation (Cambridge University Press, Cambridge, Eng., pp. xxxviii 
+ 930, 1939), of which a new edition in two handier volumes is now 
available. Examples of useful books on other, mainly temperate, lands 
include L. S. Berg's Natural Regions of the U.S.S.R. (Macmillan, New 


York, transl. edition, pp. xxxi -f- 436, 1950) and those published in Die 
Vegetation der Erde — such as J. W. Harshberger's Phytogeographic Survey 
of North America (Engelmann, Leipzig, pp. lxiii + 790 and map, 191 1) 
and L. Cockayne's The Vegetation of New Zealand, second edition 
(Engelmann, Leipzig, pp. xxvii -f 456 and additional illustrations, 1928). 
As examples of what may with advantage be done in elucidating a single 
if generalized type of vegetation in one important region, we may cite 
E. L. Braun's Deciduous Forests of Eastern North America (Blakiston, 
Philadelphia & Toronto, pp. xiv + 596 and map, 1950) and J. E. Weaver's 
North American Prairie (Johnsen, Lincoln, Nebr., pp. xi + 348, 1954), 
and, as a study of a particular ecological aspect wherever it may crop up, 
V. J. Chapman's Salt Marshes and Salt Deserts of the World (Leonard 
Hill, London, pp. xvi + 352 + index, in press). 

Whereas the world's longest plants are probably some lianes of the 
tropical rain forest which are reputed to exceed 655 feet (200 metres, cf. 
p. 431) in length (and hence not to be rivalled by the giant Pacific Kelp 
Macrocystis pyrifera, at least according to recent accounts — cf. p. 535), it 
is in the temperate regions that there grow what appear to be the world's 
tallest plants — see p. 63. For this proud title there is considerable 
doubt about the validity of claims of the past and some even about con- 
tentions of the present, but the oft-quoted and apparently well authenti- 
cated 364 feet cited on p. 63 as that of the tallest living Coastal Redwood 
no longer stands, as the tree has lost its top and is now only 346 feet high. 
News of this unfortunate loss has arrived as the present volume is in the 
press, and, at the same time, from Dr. Lincoln Constance, of the Uni- 
versity of California at Berkeley, details of another tree of Sequoia semper- 
virens, growing in the Bull Creek Flat area, that is reported to be 368-7 
feet high, though he stresses that this measurement has not been finally 
authenticated. Whereas even this height was apparently exceeded by 
some Eucalypts growing is southeastern Australia in fairly recent times, 
where heights of up to 500 feet are widely cited and one of 375 feet for a 
specimen of Eucalyptus regnans seems to be well accepted (cf. A. R. 
Penfold and J. L. Willis's Eucalyptus : Botany, Cultivation, and Utiliza- 
tion, Leonard Hill, London, in press, and J. L. Willis in litt.), there do 
not appear to be among standing trees any very close rivals of the tallest 
Sequoia sempervirens. Moreover, as the tallest known living Eucalypt 
(growing in Tasmania, and also of the so-called ' Mountain-ash ', E. 
regnans) was only 322 feet high in June, 1956, it is likely to be a good many 
years before the Australians can again rival their American cousins in 
the possession of the world's tallest living tree. 

Chapter XIII 


The Arctic and Antarctic make up the ' polar lands ' and are 
roughly those lying, respectively, north or south of the limit of 
arborescent growth, even as the high altitudes here considered are 
those above the timber-line. This ' tree-limit ' approximately 
coincides with a mean temperature of io° C. (50 F.) for the warmest 
month (normally July in the Arctic). Actually, though use of the 
tree-limit is a great improvement on the astronomically determined 
but biologically misleading arctic and antarctic circles, satisfactory 
delimitation of the Arctic and Antarctic can scarcely rest on such 
a simple basis ; nevertheless for our present purpose it will suffice to 
tell us approximately what to consider in this chapter, and what 
to exclude. 

Beyond the stunted ' elfin wood ' and twisted ' Krummholz ' that 
top the upper timbered slopes of mountains in forested regions, or 
the usually more open ' taiga ' that terminates the poleward limit of 
forests, is normally a zone of relatively luxuriant ' tundra '. This, 
by definition, is treeless, though in its most southerly tracts it may 
contain shrubby examples of the forest dominants as well as, in 
favourable situations there and elsewhere, bushes of other sorts. 
The great and extensive examples are afforded by the Arctic, the 
Antarctic having relatively little ice-free land apart from scattered 
islands, and the high-alpine tracts being also limited. Consequently 
most of this chapter is concerned with the Arctic, followed by 
briefer mention of some antarctic and high-alpine features which 
are often comparable — without, however, being by any means 

If we recognize some modern refinements 1 the Arctic may be 
generally characterized as treeless, with the winters largely dark and 
cold and the mean temperature of the warmest month plus one-tenth 
of the mean of the coldest month over a cycle of years not more 

1 Auggested, for example, in the Journal of Ecology, vol. 39, pp. 308-315, 




than 9° C, with less than fifty days between spring and fall frosts, 
with the subsoil in most places permanently frozen, with an annual 
precipitation normally below 50 cm. (commonly and widely below 
25 cm.) and largely in the form of snow which drifts and is packed 
tightly by the wind, with the soil generally moist in the summer, 
and with sheltered salt as well as fresh water frozen over during 
much of the winter. Both arctic and high-alpine regions typically 
exhibit marked microhabitat effects and consequent variability from 
spot to spot. Their slopes may also undergo ' solifluction ', which 
is a slow flowing or creeping downwards of the comminuted surface 
material over a frozen or other hard substrate, while in flatter areas 
the sorting in relation to frost action of the surface soil into various 
kinds of ' polygons ', most often with the finer material in their 
centres, is extremely widespread especially in the Far North. 

In spite of its treelessness and generally dwarfed nature, giving, 
to the layman, an impression of monotonous sameness, the vegeta- 
tion of arctic regions varies very markedly from place to place. 
This variation is often extreme in closely contiguous areas of different 
habitats or, it sometimes seems, without involving any marked 
difference in conditions — even suggesting that repeated readjust- 
ments to disturbance outweigh any tendency to equilibrium. Indeed 
one of the most striking features of arctic vegetation is its extreme 
variability from one small area to the next — in the absence of 
sufficient growth to control the physical conditions of the environ- 
ment, which conditions themselves often vary rapidly and even 
drastically from spot to spot. Thus whereas in a forest, for example, 
the vegetation largely determines habitat conditions (including the 
microclimate), in the Arctic the vegetation is relatively impotent. 
Here the struggle of plants tends to be with the inimical forces of 
a harsh physical environment rather than with hostile competitors 
as in more favourable situations, though there is still plentiful 
competition between plants in the more favourable arctic habitats. 
Such competition is particularly rife where growth is relatively 
luxuriant towards the southern limit of the Arctic ; similarly in 
high-alpine regions it is found mainly towards the lower limits, as 
in the Antarctic towards the northern boundary. 

In the arctic regions land is ranged practically around the North 
Pole, though unlike the situation in the southern hemisphere there 
is none at the very highest latitudes. In spite of considerable 
differences in flora, especially at the lowest latitudes of what we 
recognize as the Arctic, the over-all picture of vegetation is closely 


similar in the different sectors into which the Arctic may conveniently 
be divided (see Fig. 46). Thus the vegetation developed under 
similar habitat conditions in any particular climatic belt ranged 
around the top of the globe tends to look much the same in whatever 
sector it may lie, and there do not seem to be any major subsidiary 
regions that can be singled out, such as the Mediterranean or various 
semi-deserts in the temperate zone. 

Under the prevailing cool conditions, water is very widely 
sufficient in the Arctic for such limited growth as the climate, etc., 
allows, and the main vegetational differences in any particular 
belt are rather in accordance with the actual habitats (such as were 
described in Chapter XI). Thus local edaphic or physiographic 
variations can ring the most immediate and fundamental changes 
in the local plant life. On the other hand a progressive and almost 
regular over-all depauperation of the vegetation is to be observed as 
we go farther and farther north ; and as this tends to be rather 
closely comparable in the various sectors, it is deemed expedient to 
separate each sector (and consequently the Arctic as a whole) roughly 
into three main belts. These are the low-Arctic, in which the vegeta- 
tion is continuous over most areas, the middle-Arctic, in which it is 
still sufficient to be widely evident from a distance, covering most 
lowlands, and the high-Arctic, in which closed vegetation is limited 
to the most favourable habitats and is rarely at all extensive. 1 The 
following outline account of the main vegetational types of the 
Arctic will accordingly have, under each major heading, some con- 
sideration of the expression of this type in each of these three belts, 
ranging from south to north. Examples of low-arctic lands are the 
southern portions of almost all sectors, of middle-arctic lands Jan 
Mayen Island and the vicinity of Point Barrow, Alaska, and of high- 
arctic lands the whole of the Spitsbergen Archipelago, and the 
Canadian Eastern Arctic north of Lancaster Sound. 

Arctic Tundras 

The term ' tundra ', meaning essentially a treeless plain, has been 
used in so many and often such vague senses that it seems desirable, 
if we are to retain it at all, to limit its use so that it will have a more 
precise scientific connotation. In the present work the tundra 
proper is understood as the usually ' grassy ' formation lying beyond 
(or in some extra-arctic places forming patches within) the limit of 

1 See Frontispiece for a first, tentative attempt to delimit these three arctic helts. 


arborescent growth — except where shrubs or undershrubs pre- 
dominate (in scrub and heathlands), or vegetation covers less than 
half of the area (in ' fell-fields ' and ' barrens '). Although this still 
includes such special cases as salt-marshes and manured areas, it 
is customary to consider these separately, as is done in the present 
work. On the other hand, it excludes the taiga and at least the 
forested parts of the mixed ' forest-tundra ' described in the last 
chapter (p. 348). Generally comparable types occur in antarctic 
regions where, however, suitable land areas are relatively small. 
Alpine tundra bears a similar relationship to the timber-line on 
mountains. Instead of true Grasses which, however, are rarely 
absent, grass-like plants such as Sedges (Carex spp.), Cotton-grasses 
(Eriophorum spp.), Rushes (Juncus spp.), and Wood-rushes (Luzula 
spp.) commonly afford most of the ' grassiness ' of the tundra, 
though various perennial forbs are usually associated, as are often 
a sprinkling of dwarf woody plants. 

Even in this restricted sense the tundra developed in almost any 
arctic region is usually very variable, different areas supporting 
widely different types. The variation takes place particularly with 
differences in exposure and in water and other soil conditions, and, 
at all events in low- and middle-arctic regions, affords faciations 
far too numerous even to mention here. We may, however, dis- 
tinguish and outline, besides a general central type, the tundras of 
damper depressions on the one hand and of drier exposed areas on 
the other. 

The general run of tundra which covers a large proportion of the 
lowland plains and some less extensive upland areas of most low- 
arctic regions is commonly a rather thin ' grassy ' sward dominated 
by mesophytic Sedges such as the Rigid Sedge (Carex bigelowii agg.) 
and Grasses such as the Arctic Meadow-grass (Poa arctica s.l.), with 
various associated forbs and under-shrubs including dwarf Willows. 
The whole forms a continuous if often poor sward commonly 15- 
35 cm. (approximately 6-14 in.) high in which a mixture of various 
Bryophytes and Lichens usually forms a rather poorly marked second 
layer a very few centimetres high. 

Commonly the low-arctic tundra is a mosaic made up of faciations 
having each some lesser number of the total association dominants, 
and including consociations having only one of these. The areas 
of the component communities are often small and the variation 
from spot to spot in the tundra is accordingly usually considerable. 
In addition there are often local societies dominated by species other 




than the association dominants. The (sometimes unaccountable) 
mixing and even intergradation of all these communities is often 
intricate and may be suggestive of their relative youth, many having 
apparently failed to come to a state approaching equilibrium with 
the environment since emergence from glaciation or other extreme 
disturbance. Actually, it may be questioned whether, in many areas, 
even relative equilibrium can be attained in the face of the persistent 
frost-activity, and it has been claimed that the whole system con- 
stitutes an ' open ' one in which the main tendency is repeated 


-Tundra on Southampton Island, Hudson Bay. 
support low bushy Willows. 

The depressions 

readjustment to almost perpetual disturbance. Fig. 109 shows an 
area of low-arctic tundra in eastern Canada. 

With the generally poor drainage resulting from the soil being 
permanently frozen to not far beneath the surface, damper depres- 
sions or marshy open tracts tend to be plentiful although often of 
quite limited extent ; indeed they are rarely absent except in regions 
of porous substrata and low water-table. In the low- Arctic they 
are commonly rather luxuriantly vegetated, the sward often being 
taller than it is in drier areas. They are usually dominated by 
Cotton-grasses and relatively hygrophytic Sedges such as marshland 
ecads of the Water Sedge (Carex aquatilis agg.), and by Grasses such 
as the Arctagrostis (Arctagrostis latifolia s.l.), with a few hygro- 




philous Willows or other ground-shrubs and many hygrophilous or 
ubiquitous forbs. Typical among these last are Viviparous Knot- 
weed {Polygonum viviparum) and the bright-flowered Yellow Marsh 
Saxifrage (Saxifraga hirculus agg.). The fairly luxuriant cryptogamic 
layer is largely composed of Mosses, and helps to consolidate the 
whole. An example is seen in Fig. no. Often these marshy areas 
are beset with small hummocks commonly about 25 cm. high, and 
introducing drier conditions on their tops, which may then support 
heathy plants and Lichens. Such hummocky tracts are known as 

VahJaW * 

*y$^ *, 

-"■.., . * <5;;, :; . .... ; ■". r. ■'•' * ' 

\- ■■■-.:<'%■:■: 

Fig. 1 10. — Marshy tundra near the south shore of Hudson Strait, dominated by 
Cotton-grasses, Sedges, and Grasses, with dwarf Willows creeping among the 

subdominant Mosses. 

' hillock tundra '. In other instances tundras, especially of the 
damper types, are liable to be much interrupted by various of the 
geodynamic influences prevalent in cold regions — such as, par- 
ticularly, solifluction and ' patterned soil ' (polygon — see p. 381) 
formation of various kinds. 

The drier tundras of raised areas or well-drained surface material 
in low-arctic regions tend to be much poorer and thinner than the 
damper types. Typically they are composed of an extremely 
various array of more or less xerophilous Sedges (such as the Rock 
Sedge, Car ex rupestris), Willows (particularly the Arctic Willow, Salix 
arctica s.l.), Grasses (such as Alpine Holy-grass, Hierochloe alpina), 

3 86 



Northern Wood-rush (Luzula confusa agg.), and various forbs (such 
as the same Viviparous Knotweed), in addition to Mountain and 
Arctic Avens (Dryas spp.), which are somewhat woody, and which 
may dominate considerable areas. But although scattered heathy 
plants occur in them, these areas are scarcely heaths, any more than 
are the lichen-rich ones dominated by xerophilous Sedges that 
characterize dry and exposed situations (see p. 392). Moreover 
their vegetation is usually rather poor, often barely covering the 

Fig. hi. — Dry tundra on raised area overlooking Hudson Bay, composed princi- 
pally of an intricate mixture of xerophilous Lichens, Grasses, Sedges, and other 
herbs. Dwarf woody plants also occur, and the surface is interrupted by pro- 
jecting lichen-covered boulders. 

ground in spite of a plentiful admixture of Lichens and sometimes 
also of Bryophytes. Fig. 1 1 1 shows such an area in which boulders 
project through the thin and somewhat heathy, lichen-rich vegetation. 
Especially on limestone or porous sandy substrata is growth often 
poor and the vegetation relatively sparse, although the component 
flora particularly in calcareous areas may be very various. 

It would accordingly seem that the major variations in the precise 
type of tundra take place chiefly, but by no means solely, with 
local water conditions working through exposure or edaphic factors, 


while very locally the effect of frost action may be paramount. 
Thus differences in substratum, as between limestone and acid- 
weathering rock, can introduce vegetational differences due to 
particular plants' preferences quite apart from water-relations, while, 
as an example of the entry of another factor, heavy pasturing can 
lead to increased grassiness as in temperate regions. In addition, 
polygon-formation and solifluction may cause persistent disturbance. 
It may be noted that, whereas the dominants are usually at lea^t 
specifically distinct in different types of low-arctic tundra, some of 
the less exacting, more tolerant associates may be present in a wide 
range of habitat types. This again is comparable with the situation 
in cool-temperate regions and, it often seems, obtains still more 
forcibly to the north. Thus in the Far North some of the hardier 
plants, such as Viviparous Knotweed and some of the Saxifrages, 
grow in an extraordinarily wide variety of habitats, ranging from wet 
to dry, exposed to sheltered, and open-soil to. vegetationally ' closed \ 

The middle-arctic belt is characterized by tundras of a generally 
poorer type, both in the matter of flora and luxuriance of develop- 
ment, than the low-arctic ones. Thus some of the plants which 
were important in low-arctic tundras are absent, though all of the 
dominants, etc., mentioned above for low-arctic tundras can, and 
frequently do, occupy a similar position also in middle-arctic regions. 
Moreover the range of types is much the same, damp, mesophytic, 
and drier ones being distinguishable. An example of the second, 
dominated by mixed Grasses and Sedges, in northernmost Alaska 
overlooking the Arctic Ocean, is shown in Fig. 112, from which it 
may be seen that growth tends to be lower and poorer than in low- 
arctic regions, though this particular area is only just middle-arctic 
in type. 

In high-arctic regions still further depauperation is general, and 
indeed only limited and relatively few areas are sufficiently vegetated 
to be designated as tundra. These areas are chiefly marshy ones 
and may be still dominated by Sedges, Cotton-grasses, and Grasses 
— often of the same species as in the South, and including similar 
associated forbs, though woody plants apart from prostrate Willows 
are usually absent. Mosses commonly consolidate the whole, and 
in some places appear to dominate. Fig. 113 shows an unusually 
extensive area of marshy tundra in Spitsbergen, characterized by 
peaty hummocks up to 25 cm. high, and of a type commonly termed 
' hillock tundra '. While the main dominants in such areas are 
commonly Sedges and Grasses growing on the sides or tops of the 

3 88 




Fig. 112. — Mesophytic Sedge-Grass tundra on the north coast of Alaska near 
Point Barrow, overlooking the Arctic Ocean. Ice-pups are seen stranded on the 
shore, and in the distance the margin of the polar pack-ice is visible. The tundra 
in flat areas is usually closed; but on slopes, where disturbed by geodynamic 
influences, it is commonly discontinuous. 

FiG. 113. — Extensive area of damp ' hillock tundra ' on the coast of Spitsbergen. 
Note the grassiness of the hummocks but frequent puddles of water between. 




hummocks, the microhabitat effect is extreme, the microhabitats 
ranging from depressions occupied by dark boglets or puddles of 
'free' water (seen in Fig. 113) to dry hillock tops occupied by 
Lichens or, in favourably sheltered situations, xeromorphic ground- 

Tracts of ' grassy ' mesophytic tundra of any substantial extent 


Fig. 114. — Discontinuous tundra-like tract of mixed Grasses, Northern Wood- 
rush (Luzula confusa agg.), forbs, and Polar Willow (Salix polaris agg.), in inland 
valley, West Spitsbergen, grazed by a pair of wild Reindeer. Beyond is the dry 
bouldery bed of a melt-water stream that is dry during most of the summer, and, 
behind, barren scree and other slopes. 

are not common in the high- Arctic, though Fig. 114 shows a dis- 
continuous but tundra-like patch of mixed Grasses, Northern Wood- 
rush (Luzula confusa agg.), forbs, and Polar Willow (Salix polaris 
agg.), that is sufficiently developed to attract Reindeer. It is situated 
in a sheltered valley well inland in Spitsbergen, by the side of a 
bouldery bed of a snow-water stream that is dry during most of 
the summer. Still drier types of tundra in these farthest north 
lands tend to be dominated largely by Lichens and to be much 
interrupted by rocks or bare patches — especially in exposed situations. 


So far as regular ecological successions' are concerned, these are 
especially problematical in the Arctic. It has, however, been sug- 
gested that the marshy and dry tundras may be subclimax and the 
mesophytic ones climax or preclimax, the scrub and heathlands, 
which are developed in the most favourable situations (see below), 
being either postclimax or, perhaps, indicative of a more general 
climax to be expected ultimately in sufficiently favourable situations, 
though at present a mixed ' polyclimax ' is commonly found. The 
significance of different ' stages ' in the hypothetical successions 
may, however, vary from place to place. Thus, in the Far North, 
heathy plants are apt to be so restricted to the most favourable 
situations as to suggest that without major climatic change they 
could not become widely dominant in the manner already obtaining 
in some places in the low-Arctic. Moreover, frost and other dis- 
turbance is so widespread, inter alia impeding or even preventing 
the maturation of soils, that it seems as though many areas undergo 
a kind of perpetual readjustment rather than exhibit the tendency 
to equilibrium which is implicit in a real climax. 

Arctic Scrub and Heathlands 

A shaggy scrub of Willows and/or Birches is commonly developed 
on the most favourable slopes, in damp depressions, and especially 
along watercourses and the margins of lakes in low-arctic regions. 
It is commonly around 60 cm. (about 2 feet) high, as in the example 
shown in Fig. 115, but tends to become lower and more restricted 
northwards until, about the centre of the middle-arctic belt, it 
becomes usually very limited in extent and stature. However, in 
the most favourable situations in the extreme south the Willows 
may be luxuriant (cf. Fig. 116) and even exceed the height of a 
Man, and especially in southwestern Greenland the scrub is quite 
extensively developed, in some places including arborescent Birches. 
These Greenland Birch ' forests ' are of very limited extent, with 
the trees scattered and scraggy though sometimes nearly 6 metres 
in height and 25 cm. in stem diameter. Their areas have been 
termed subarctic but seem too limited to separate on an over-all, 
world basis ; they are also too fickle, the development of an arbores- 
cent habit being evidently dependent on local shelter, etc. Apart 
from these larger Birches, the main dominants in different regions 
are most often the Dwarf Birch (Betula nana agg.) or Scrub Birch 
(B. glandnlosa agg.), or such shrubby Willows as the Glaucous 

Fig. 115. — Tangled Willow scrub in the low-arctic belt of the Northwest Ter- 
ritories, Canada. The scrub occupies a slight depression whose depth is indicated 
by a spade resting on the ground (in centre). 

Fig. 116. — Patchy scrub of Glaucous Willow (Salix glauca s.l.) and Scrub Birch 

(Betula glandulosa agg.), up to nearly 2 metres high, in southwestern Greenland. 

The scrub is interrupted by grassy tracts of fair turf which appear to result from 

ancient pasturing {see page 222). 

39 1 


Willow (Salix glauca s.L), the Broad-leafed Willow (S. cordifolia s.l.), 
the Feltleaf Willow (S. alaxensis agg.), or Richardson's Willow (S. 
richardsonii agg.). Often two or more of these shrubs will dominate 
a mixed association. In some places bushes of Green Alder (Alnus 
crispa agg.) are present and may be locally dominant. 

Such scrub at its best is so thickly tangled and produces so much 
litter that few associated plants occur, apart from tall Grasses such 
as Bluejoint (Calamagrostis canadensis agg.) and occasional straggling 
forbs. But where the dominants are less luxuriant, an extensive 
flora is often found, including a considerable variety of herbs and 
Mosses, or, in dry situations, subdominant heathy plants such as 
Crowberry (Empetrum nigrum s.L). Also characteristic of dry scrub 
are patches of tall Cladonias, Stereocaulons, and other Lichens, with 
or without Polytricha or other coarse Mosses. To the north such 
scrub thins out gradually, its most northerly expression about the 
northern limit of the middle-arctic belt being usually in the form 
of single or scarcely confluent bushes that rarely exceed 50 cm. in 
height and are usually much lower, though often quite wide. 

Heathlands are more widespread and various in the Arctic than 
is scrub, though still commonly occupying only a very small propor- 
tion of the total area. They are usually characterized by being 
dominated by members of the Heath family (Ericaceae) or by heath- 
like plants such as, particularly, Crowberry. Sometimes, however, 
broad-leafed plants such as Avens (Dryas), or Sedges such as the 
Nard Sedge (Carex nardina s.l). or Bellard's Kobresia (Kobresia 
myosuroides), may dominate dry and usually exposed, lichen-rich 
areas that are often classed as heathlands rather than among the 
drier tundras with which they seem more properly to belong (see 
p. 386). Leaving aside such cases it may be said that heathlands 
in the Arctic tend to be confined to the more favourable, sheltered 
situations that are snow-covered in winter — provided they are not 
too moist in summer. In many regions they characterize coarse- 
grained rather than clayey soils, as pointed out by Professor Thorvald 
Sorensen (in lift.). 

In the low-arctic belt the heathlands are usually covered by a 
continuous thick sward of mixed woody and herbaceous plants, the 
main dominants being typically 8-15 cm. high. These commonly 
include Crowberry, Arctic Blueberry (Vaccinium uliginosum subsp. 
alpinum), Mountain Cranberry (V. vitis-idaea agg.), Arctic Bell- 
heather (Cassiope tetragona), Narrow-leafed Labrador-tea (Ledum 
palustre agg.), Dwarf Birch, and various diminutive Willows. Often 




the dominants themselves are much mixed, and usually they are 
consolidated below by a layer of cryptogams in which Mosses or 
Lichens commonly subdominate according to whether the situation 
is relatively moist or dry, respectively. Fig. 117 shows an area of 
dense mixed heath on the south shore of Hudson Strait. In the 
drier situations there may occur frequent gaps in the heath which 
are actually dominated by Lichens — particularly by ' Caribou-moss ' 
Cladonias that may form a sward 5 or more cm. high. In depres- 
sions and behind obstructions where snow drifts deeply in winter, 

Fig. 117. — Dense low-arctic heath dominated by Arctic Blueberry (Vaccinium 

uliginosum var. alpinum) in northernmost Quebec. Light-coloured Lichens and 

leaves of dwarf Willows and Sedges are visible. To the left of the sheath-knife 

is a flowering bushlet of Lapland Rose-bay (Rhododendron lapponicum). 

a characteristic dark (except when flowering) heath dominated by 
Arctic Bell-heather usually develops, often with associated Sedges 
and Mosses at least where the soil is lastingly moist. Such an area 
is shown in Fig. 118 and, apart from a zone of more mixed heath 
that may develop outside, usually constitutes the outermost of the 
zoned series of subclimaxes developed in late-snow areas as described 
on pages 402-5. 

In the middle-arctic belt, heathlands are usually somewhat lower 
in stature and more restricted in area than to the south, having the 
appearance of postclimaxes developed in the most favourable situa- 
tions. Of the cited dominants Mountain Cranberry has usually 


i i 

Fig. i i 8. — 'Snow-patch' darkened by Arctic Bell-heather (Cassiope tetragona), 
southern Baffin Island. 

FlG. i 19. — Mixed middle-arctic heath with many light-coloured and other Lichens. 
Bushlets of Narrow-leafed Labrador-tea {Ledum palustre agg.) are seen to the 
right and left of the pipe, which is 15 cm. long and gives the scale. Northern 

Baffin Island. 


disappeared, and although the taller ones may still exceed 20 cm. 
in height the sward is usually only 5-10 cm. high. Whereas it may 
still be fairly dense, more often the ' heath ' is of scattered ground- 
shrubs with intervening thin patches of Cetrarias, Alectorias, and 
other Lichens as seen in Fig. 119. 

In the high-arctic belt heathy plants are entirely absent over 
considerable areas, the tracts that are popularly spoken of as ' heaths ' 
being usually dominated by Avens, Sedges, or even Lichens. How- 
ever, Crowberry or Arctic Blueberry plants are to be found in some 
regions, dominating limited heathy communities in unusually favour- 
able situations, while Arctic Bell-heather is quite widespread, char- 
acteristically forming a dark tract where the snow accumulates 
sufficiently to form a good protective covering in winter (though 
disappearing early in the growing-season). 

Arctic Fell- Fields and Barrens 

These are types in which the evident vegetation occupies less 
than half of the area ; and whereas the two categories are scarcely 
to be rigidly distinguished, it is usually those tracts that bear rela- 
tively few and scattered plants that are referred to as ' barrens '. 
Fell-fields typically have a surface of frost-shattered detrital material 
including much finer ' soil ' and usually support fairly numerous 
different species forming mixed communities, whereas barrens are 
apt to be characterized by one prominent size of particle and a 
single species of plant, such as Mountain Avens or Purple Saxifrage 
(cf. Fig. 128). This is especially the case when they occupy the 
most exposed situations. 

Where sufficient moisture is present these poorly-vegetated areas, 
like some tundras, are commonly disturbed by all manner of frost- 
heaving and allied effects — such as solifluction on slopes and polygon- 
formation on the flatter terrain. The solifluction is generally 
manifest in streaks extending longitudinally downhill, adjacent 
streaks being either of different material or accentuated by vegeta- 
tion which cannot grow on the more dynamic parts. The ' polygons ' 
are almost endlessly variable but very commonly have the form of 
polygonal or circular areas f-2 metres in diameter and composed of 
finely comminuted soil that is apt to be too dynamic to support 
any plants at all, separated by narrow intervening tracts containing 
most of the larger stones and often raised and vegetated or in other 
cases barren (Fig. 120). Or the polygons may be separated by 


narrow cracks or troughs that afford shelter for plants which in the 
Far North often grow better in such microhabitats than in surround- 
ing areas. 

On steep slopes and below weathering crags, still more dynamic 
and often poorly vegetated ' screes ' are common. Also often con- 
stituting barrens of one sort or another are ' raised beaches ' near 
sea-shores ; for although some are well vegetated, many others are 
the reverse, owing to exposure, recent emergence, or an unfavourable 

Fig. 120. — ' Polygons ' in northernmost Spitsbergen. The stony intervening 
tracts are here almost barren, but often in other instances are covered with 


substratum. Altogether these poorer types of vegetation — or ter- 
rain, for often plants are scarcely at all in evidence — are so numerous 
and variable in the Arctic that only a few examples can be mentioned 

In low-arctic regions fell-fields, barrens, and the like, are found 
chiefly in upland districts and in exposed areas near the coast — 
especially where the substratum is of porous material. Here, owing 
to lack of stability, to local aridity, or to extreme exposure, such 
fell-field or ' half-barren ' areas as that shown in Fig. 121 occur, 
in which Mountain Avens, Nard Sedge, and various tufted Saxifrages 
and other herbs form irregular patches of vegetation. A fair number 
of cryptogams are often admixed, though usually they are of poor 




growth. In the most unfavourable situations of all, this type may 
thin out to a stony barren supporting little more than diminutive 
crustaceous Lichens and very occasional depauperated tussocks of 
Avens or Purple Saxifrage. 

Although such poorly-vegetated areas tend to be more numerous 
and widespread in the middle-arctic zone than farther south, they 
still do not normally occupy the general run of lowland terrain but 
are chiefly encountered in exposed situations (as for example in the 

Fell-field on calcareous soil in exposed situation, northernmost 

foreground of Fig. 128). An extensive example is shown in Fig. 
122, from an altitude of 671 metres (2,200 feet) in central Baffin 
Island, in which Lichens of poor growth cover much of the surface, 
vascular plants being virtually absent. 

In most high-arctic regions ' open ' and often extremely sparse 
vegetation is the general rule, and so fell-field and barrens areas are 
widespread and plentiful. A relatively well-vegetated and extensive 
area, reminiscent of many observed by the author in far northern 
Ellesmere Island and Spitsbergen, is seen in Fig. 123, which he 
took, however, in the vicinity of the North Magnetic Pole on Prince 
of Wales Island. It shows a monotonous expanse of mixed but 
scattered and open, diminutive herbs and terricolous (i.e. earth- 
inhabiting) Lichens, with occasional small tufts of Avens (in this case 
Dry as integrifolia agg.). The general aspect is desolate in the extreme 




Fig. 122. — Lichen barrens in the uplands of central Baffin Island, looking south. 
Note the virtual absence of higher plants and the abundance of persistent snow- 
patches on north-facing slopes. 


' if- 


Fig. 123. — Monotonous tract of prairie-like fell-field in the vicinity of the Magnetic 
North Pole, Prince of Wales Island, Canadian Arctic Archipelago. The sparse 
vegetation consists of scattered diminutive herbs, terricolous Lichens, and occa- 
sional tufts of Arctic Avens, all growing in open formation. 


in most of such exposed high-arctic tracts, with a large proportion 
of the area typically occupied by barrens supporting little more than 
occasional tufts of Arctic Poppy (Papaver radicatum s.l.) or Purple 
Saxifrage (Saxifraga oppositifolia agg.), although more mixed fell- 
fields may occur in less unfavourable situations. On well-drained 
banks there may be Grasses and some hardy but attractive forbs, 
while very occasionally under the most favourable conditions a 
limited tract of thin heath may be developed, though often one can 
trek for days without encountering such a manifestation. The only 
at all extensive tracts of closed vegetation are mostly of rather thin 
marshy and often hummocky tundra (Fig. 113), and even here the 
dominants rarely exceed a height of 30 cm. above the surface from 
which they grow. Yet most of these vegetation-types and a fair 
range of vascular plants are to be found right up to the highest 
latitudes of land, between 83 and 84 N., in marked contrast to the 
situation in Antarctica (see pp. 415 et seq.). 

Seaside and Other Local Types 

On sandy and fine-shingle maritime beaches in both low- and 
middle-arctic belts there are usually scattered plants of Sea-purslane 
(Arenaria peploides agg.) and Sea Lungwort (Mertensia maritima 
agg.) on the foreshore and, farther up, stabilizing beds of Lyme- 
grass (Elymus arenarius s.l.) which may be fairly tall and luxuriant 
(Fig. 124). In the high-arctic belt, Lyme-grass is unknown and 
the other two are rare, so exposed sandy and shingly shores are 
liable to be barren around high-tide mark. 

In sheltered and less well-drained seaside areas, muddy or sandy 
' salt-marshes ' are common though usually of very limited extent 
in the Arctic. Even more than many other types of vegetation, 
they show close similarity of form all around the southern portions 
of the Arctic — and also, with natural depauperation, far northwards. 
Thus in low-arctic areas they typically consist of a dwarfish grassy 
sward dominated by Alkali-grasses (particularly the Creeping Alkali- 
grass, Puccinellia phryganodes agg.) and Sedges (particularly the Bear 
Sedge, Carex ursina, and phases of the Salt-marsh Sedge, C. salina 
s.l.), with associated Low Chickweed (Stellaria humifusa), Scurvy- 
grass (Cochlearia officinalis s.l.), Pacific Silverweed (Potentilla egedii 
agg.), and other halophytes. Fig. 125 shows such a salt-marsh 
on the south coast of Baffin Island. Except for the usual absence 
farther north of the Silverweed and the substitution of Salt-marsh 



Fig. 124. — Shingly beach bound by swarded Lyme-grass (Elymus arenarius s.l.). 

Farther down, between tide-marks in this sheltered inlet, the shore is darkened 

by algal growth. Hudson Strait, northeastern Canada. 

Fig. 125. — Looking down on a salt-marsh dominated by Pacific Silverweed 
(Potentilla egedii agg.) (flowering) and Creeping Alkali-grass (Puccinellia phryganodes 
agg.) (light-coloured stolons). Pipe gives scale. South coast of Baffin Island. 




Sedge by the doubtfully specifically distinct Hoppner Sedge (Carex 
subspathacea), the same plants generally play a similar role in middle- 
and high-arctic regions, though with increasing depauperation. It 
seems probable that they are unable to alter their habitat markedly, 
at least if and when it has reached the approximate level of the 
highest tides, and consequently that they represent a subclimax 
which will persist indefinitely. 

Another type of local climax is engendered by the perennial 
manuring that takes place around the ' bird-cliffs ' where countless 

Fig. 126. — Luxuriant ' patchwork quilt ' of mixed and many-coloured Lichens 
and Mosses developed near top of bird-cliff. Northernmost point of Quebec, 
overlooking Hudson Strait. Scale indicated by pack on left which is 60 cm. high. 

sea-birds nest every summer. Here unoccupied ledges may support 
coarse Grasses and rank Scurvy-grass, the rock faces being covered 
by Lichens, often of extraordinary size. The tops of the cliffs 
typically support near their edges and in damp depressions a rich 
grassy sward, and, stretching back for 100 metres or so, a luxuriant 
and dense ' patchwork quilt ' of mixed and variously coloured Lichens 
and Mosses. This is seen in Fig. 126 and is due to manuring 
effects engendered apparently largely by scavenging Sea-gulls and 
Birds-of-prey, though in some instances Foxes and Polar Bears are 
also involved. The situation being usually very exposed, the 
adjacent unmanured cliffs and hinterland are apt to be practically 
barren and in striking contrast. More local manuring may also 




give remarkable effects, such as the grassy' or flower-decked swards 
that develop around human habitations, mammalian burrows, or, 
particularly, the nesting-grounds of Geese, Eiders, and other 
gregarious wildfowl (Fig. 127). An instance of a different kind 
is shown in Fig. 128, in which a dense grassy patch is developed 
around a boulder in an exposed Purple Saxifrage barren overlooking 
the sea ; to such prominences Birds and predators repair, manuring 
the ground in the immediate vicinity and doubtless sometimes 

Fig. 127. — Looking down on a luxuriant mossy mat on the manured periphery 
of a wildfowl nesting-ground in Spitsbergen. The plant flowering on the right, 
above the matchbox (giving scale), is Yellow Marsh Saxifrage (Saxifraga hir cuius 
agg.), the flowers on the left are of Alpine Brook Saxifrage (S. rivularis agg.), 
and the small ones below are of Fringed Sandwort {Arenaria ciliata s.l.). 

bringing in viable seeds. The result is often a luxuriant if limited 
sward in an otherwise seemingly sterile situation. 

As very little cultivation or other human disturbance has so far 
taken place in most arctic areas, few tracts bear witness to such 
change ; on the other hand a common and widespread type of local 
climax is that engendered by the drifting and late-melting of snow. 
This tends to take place similarly each year and to lead to char- 
acteristic vegetational zonation within the area of the drift. The 
zones produced are of subclimax nature although the outermost may 
be considered postclimax — especially in the Far North where the 




most widespread and often the sole heathy plant, Arctic Bell- 
heather, may be practically confined to such situations. In the low- 
and middle-arctic belts the outermost zone of such ' snow-patches ', 
which is well protected in winter by snow but does not have its 
growing-season markedly reduced by late melting, is commonly 
vegetated by a luxuriant mixed heath (Fig. 129), or, in lastingly 
damp situations, by a thin Willow scrub. Farther in, where 
the snow drifts sufficiently deeply for the growing-season to be 

Fig. 128. — Purple Saxifrage barren on exposed ridge overlooking the sea in 
northernmost Baffin Island. The Saxifraga oppositifolia agg. forms only scattered 
tufts and small dark matlets that scarcely show in the photograph; nevertheless, 
around a prominent boulder that has persistently acted as a perch, there is a 
luxuriant grassy sward where the ground has been manured. 

appreciably shortened, Arctic Bell-heather characteristically forms 
a dark belt, often being the sole dominant as in Fig. 118. In 
other instances this belt may be more mixed and only 1-3 metres 
wide, as in Fig. 129, which shows in the background the more 
barren inner zones. These are apt to vary considerably in number 
and vegetation in different places and circumstances. However, they 
typically include towards the outside a zone of dwarf Willows 
(particularly Herb-like Willow, Salix herbacea, as in Fig. 130), 
and, farther in where the growing-season is too short for woody 
plants, a sparsely vegetated zone with a considerable variety of 

40 4 



Fig. 129. — ' Late-snow ' patch in the highlands of central Baffin Island, showing 
in foreground the outer zone of mixed Arctic Blueberry heath, then a narrow belt 
of dark Arctic Bell-heather interrupted by light-coloured Lichens, and behind, 
a Herb-like Willow zone (see Fig. 130), the centre of the snow-drift area where 
the snow melts latest being occupied by a herb ' barren.' 

Fig. 130. — Looking down on the Herb-like Willow zone of the late-snow area 
shown in Fig. 129. The rounded leaves are those of the dominant Willow, the 
ground being characteristically encrusted with cryptogams. Pipe 15 cm. long 

gives scale. 


Brvophvtes and open-soil herbs such as the Mountain Sorrel 
(Oocyria digyna). 

Many of the smaller snow-patches, of course, disappear well before 
the end of summer, especially in the South, and consequently show 
only the outermost zones. On the other hand around the centre 
of the deeper drifts, which usually form in ravines, depressions, or 
behind banks or ridges, the snow may melt only towards the very 
end of summer or in a cool season not at all. Here most herbs are 
unable to persist and even cryptogams are little in evidence, though 
some tufts or limited mats of Bryophytes and investments of Algae 
are usually to be found, together with the diminutive grass Phippsia 
algida agg. (Frigid Phippsia). Towards such centres various plants, 
including attractive Saxifrages and Buttercups {Ranunculus spp.), 
are often to be found flowering at the very end of summer, sometimes 
being caught still in bud by the frosts and snow of a new winter. 
In the Far North the zones are apt to be reduced in number but 
extended in area, the fell-fields and barrens often representing inner 
zones where the modest snow-covering melts late under the pre- 
vailingly cool conditions. Here, as on mountains farther south, 
many of the snow-patches are perennial or even eternal, the chief 
growths near their centres being of Bryophytes and Algae in the 
run-off below. 

Seral Types 

Whereas the apparent arctic counterparts of many southern seral 
types have already been dealt with, being often seemingly incapable 
of leading to permanent higher vegetation or best regarded as sub- 
climax (in view either of persistent disturbance or of the extreme 
slowness of vegetable build-up), other instances of seral stages remain 
to be mentioned. Outstanding are the marshy and boggy ones of 
hydroseres (the fully aquatic communities comprising the early 
stages of which will be dealt with in Chapter XV), various ones of 
lithoseres, and the ' flower-slopes ' that probably belong to mesoseres. 

The hydrosere of arctic lakes and tarns usually has as its first 
aerial stage a ' reed-swamp ' of aquatic Sedges (particularly the 
Water Sedge, Car ex aqualilis agg.) and/or Cotton-grasses (particularly 
the Tall Cotton-grass, Eriophorum angustifolium agg.), though some- 
times Common Mare's-tail (Hippuris vulgaris s.l.), or such coarse 
Grasses as the Tawny Arctophila (Arctophila fulva agg.), may largely 
or wholly take their place. Any of these plants may form luxuriant 




beds where the bottom is soft and not more 'than about 40 em. deep. 
They generally project some 20-40 cm. above the water in the south 
but usually less in the Far North, where Mare's-tail is normally 
absent. Commonly such beds are accompanied by aquatic Mosses 
and, of course, numerous small Algae. Behind there stretches 
typically a marshy sedge-meadow with the same ' grassy ' dominants 
and, in addition, Arctagrostis and lowly Willows. This in turn 
merges into damp tundra. Fig. 131 shows such a sequence in 
the low- Arctic, though it should be noted that in exposed situations, 
especially farther north, wave action may prevent reed-swamp 
formation, a definite ■ hard line ' then delimiting terrestrial from 

Fig. 131. — Luxuriant lakeside marsh dominated by Water Sedge (Carex aqnatilis 

agg.) and Tall Cotton-grass {Eriophorum angustifolium agg.), which both extend 

out into the water. Southampton Island, Hudson Bay. 

aquatic communities. On the other hand there may be an ecotone 
of Willow scrub that, at least on its lower side, probably represents 
a later stage in the hydrosere. 

Boggy areas, typically dominated by Bog-mosses {Sphagnum spp.) 
and rather strongly acidic in reaction, are chiefly developed in the 
southern portions of the Arctic. Baked-apple (Rubits chamaemorus) 
is a characteristic inhabitant of them. Often they are well developed 
around pools in peaty tracts and are colonized by heathy plants, 
with little doubt ultimately turning into heathlands. In the Far 
North, however, many tarnside areas remain to this day uncolonized 
by higher plants — as in the right background of Fig. 132, although 
the foreground is vegetated by a fine bed of Scheuchzer's Cotton- 
grass [Eriophorum scheuchzeri). 




Lithosere stages are abundant in the Arctic, where much of the 
terrain is of more or less bare rock that has, in many cases, been 
freed from glaciation only in relatively recent times. Nevertheless 
there is no doubt that succession is proceeding, however slowly — ■ 
at all events in areas that are not too rigorously exposed or lastingly 
snow- or ice-covered. Thus rock faces, whether of glacial boulders 
or of detrital, clirT-face, or some other nature, are apt to be largely 
invested with crustaceous and foliose Lichens, and to occupy con- 
siderable areas. On the other hand, rock crevices or interstices 

Fig. 132. — Fine bed of Scheuchzer's Cotton-grass (Eriophorum scheuchzeri) beside 
tarn in northern Spitsbergen, though the waterside behind (on right) is devoid 

of higher plants. 

often support higher life-forms, so that in time a moss-mat or mixed 
herbaceous community develops, and, ultimately, heathy vegetation 
in suitable situations. 

Screes, if not too active, may also be bound by hardy plants — 
especially in low-arctic regions, where dark strips stabilized by 
vegetation often extend down scree slopes. Also commonly stabilized 
by vegetation are inland sandy areas, though the psammosere may 
advance little beyond the pioneer stage of sand-binding Mosses (such 
as Polytrichum spp.) and ground-shrubs (such as Crowberry and 
Alpine Bearberry, Arctostaphylos alpina agg.). Consolidation of the 
ground-shrubs produces already an advanced type of vegetation. On 



Fig. 133. — Top of flower-slope below weathering crag in southern Baffin Island. 
On right is seen flowering Three-toothed Saxifrage (Saxifraga tricuspidata), on 
left Arctic Fireweed (Epilobium latifolium) and Alpine Chickweed (Cerastium 
alpinwn s.l.). In the centre are two species of Fleabane (Erigeron). Knife with 
handle 12 cm. long gives scale. 

Fig. 134. — Spitsbergen flower-slope dominated by Alpine Arnica (Arnica alpina 

s.l.). The background of swarded cryptogams helps to make this high-arctic 

community look unusually luxuriant. 


both rock and sand, a dense mat of the ' silvery ' Moss Rhacomitrium 
lannginosum may cover substantial areas and apparently persist for 
many years, though in the end it is usually colonized by Lichens 
and Grasses, etc., to form what is sometimes termed ' Rhacomitrium 
heath '. 

The mesosere is represented by relatively short-lived communities 
in such favourable situations as alluvial deltas and the beds of 
receding tarns, and apparently by longer-lived types on the earthy 
or gravelly ' flower-slopes ' that form such a pleasing feature on 
steep south-facing inclines particularly in low-arctic lands. Fig. 
133 shows an example in which Fleabanes (Erigeron spp.) and 
Saxifrages are prominent, although a large variety of other forbs 
occur — typically much-mixed and in such profusion that the usual 
dominants are excluded. The writer has even encountered very 
limited communities of this type near 78 N. in Spitsbergen and 
at high altitudes in southern Greenland, growing under an unusually 
favourable combination of conditions of shelter, aspect, water, 
aeration, and soil — cf. Fig. 134. 

High Altitudes 

In general, temperatures get lower and lower as we ascend moun- 
tains, and higher and higher as we go farther and farther south at 
a particular altitude (such as sea-level) in the northern hemisphere. 
Precipitation, windiness, fogginess, and the intensity of radiation 
also tend to increase with altitude on mountains, though precipitation 
falls ofT at the higher altitudes and other conditions often interfere 
with radiation. Consequently, particular zones of vegetation in 
general get higher and higher in the mountains as we travel towards 
the Equator, though something like arctic conditions and attendant 
vegetation may still be found at very high altitudes in tropical lands. 
But although a general similarity to high-latitude lands prevails in 
high mountains even near the Equator, the light and other climatic 
regimes are by no means identical, while the floras are not necessarily 
even comparable. Nevertheless the very general (but often only 
superficial) similarity between polar regions and high altitudes else- 
where, which to some degree may extend to the vegetation, makes it 
desirable to treat high-altitude plant communities here, though 
very briefly. 

While in high-arctic lands there is the usual general tendency 
towards limitation of flora and depauperation of vegetation as 


mountains are ascended, fairly luxuriant if limited patches of more 
or less closed vegetation are to be seen in some places up to altitudes 
of at least 700 metres. Around such altitudes extensive grassy 
tundra can occur in middle-arctic regions, as it can even higher up 
in low-arctic regions — where, on the other hand, extraordinary 
barrenness may already prevail lower down. Indeed remarkably 
drastic variation is apt to be found in arctic uplands, often from 
spot to spot in closely contiguous areas. Thus while in one place 
the aspect is as of a plantless desert, in another there are plentiful 
hardy rosette and other herbs such as Saxifrages and Arctic Poppv 
(Papaver radicatum s.l.), while in favourable situations a more or 
less continuous tundra may prevail, or even a closed ' heath '. In 
general, something approaching most lowland communities persists 
well up into the mountains in the Arctic, while even near glaciers 
or perennial snow, and at quite high altitudes and latitudes, the 
climber may be agreeably surprised by a considerable show of 
vegetation including herb- or moss-mats, and even heaths or flower- 
slopes in the South. However, at very high altitudes even towards 
the southern boundary of the Arctic, extreme barrenness is usually 
encountered, with much snow and neve persisting through the 
summer wherever the local topography allows, and little else save 
crustaceous and foliose ' Tripe de Roche ' and other Lichens on the 
exposed rock faces (Fig. 135). 

In temperate regions the zone of tundra, etc., just above the 
timber-line is in many ways comparable with low-arctic regions 
near sea-level, while higher up in both instances a similar sequence 
of altitudinal climaxes prevails. Thus tundras of various types 
roughly comparable with those of the Arctic are found above the 
elfin wood to the south, with often extensive tracts of scrub (Fig. 
136, and cf. Fig. 115) where conditions are suitable and pasturing 
is not too severe. Above this may stretch increasingly limited tracts 
of ' alpine meadow ' consisting of short and more or less matted 
Grasses, Sedges, and forbs or under-shrubs (Fig. 137, A) ; heathlands 
may develop in particularly favourable situations, and fell-fields or 
barrens in detrital or exposed ones. Mossy mats are especiallv 
characteristic of run-off areas below snow-banks hereabouts. 

Still higher up and nearer the Equator there is not merely a rigorous 
climate to contend with but, also, geodynamic influences which are 
often powerful, so that conditions and the attendant vegetation 
may vary considerably from spot to spot. Almost all the vascular 
plants persisting here are chamaephytes or hemicryptophytes, and 

Fig. 135. — High in the mountains near the margin of the ice-sheet in southern 

Greenland. Valley glaciers are plentiful and patches of snow and neve (iced firn) 

persist through the summer where local topography allows. The macroscopic 

vegetation is often limited to Lichens on the exposed rocks. 

Fig. 136. — Upland scrub of Dwarf Birch and silky-leafed Willows constituting 
an altitudinal climax above tree-limit in northern Norway. 


high-alpine ' meadow 
B, plants flowering at 

Fig. 137. — High-alpine vegetation and flowering. A, a 
in British Columhia, Canada. (Phot. W. S. Cooper.) 
about 5,944 metres (c. 19,500 feet) on Mohala Bhanjyang, West Nepal: visible 
in the gravelly fell-field are Lagotis glauca s.l. (to left of English shilling giving 
scale), Potentilla saiuidersiana var. caespitosa, Pedicularis sp., Armaria sp., 2 
Leguminosae, and one each of Umbelliferae, Cruciferae, and Gramineae — also 
several Lichens. (Phot. O. Polunin.) 



a large proportion are obviously xeromorphic, being reduced in 
stature, very hairy, or otherwise modified to conserve water. As 
compared with those of lowland plants, the leaves tend to be smaller 
and thicker, with a greater development of protective tissue. Indeed 
the general aspect is much as in the Arctic, with hardy cushion or 
rosette plants including bright-flowered Saxifrages often in evidence, 
and Lichens and Mosses plentiful, any woody plants being excessively 
dwarfed. However, more and more of the actual species tend to 
be different as we travel south, until in the tropics very few arctic 
inhabitants are left, although the general aspect may still be some- 
what arctic-like at very high altitudes. Moreover, there are fre- 
quently anatomical and other differences between arctic and alpine 
plants even of very close systematic alliance. Fig. 137, B, shows a 
wide range of plants flowering at an altitude of about 5,944 metres 
(c. 19,500 feet) on Mohala Bhanjyang, West Nepal, in the Himalayas 
— higher than flowering plants are commonly supposed to go in any 
number, and perhaps constituting a record in this respect. The 
genera are often represented in the Arctic, and at least one arctic 
species, Lagotis glauca s.L, is visible. At lower altitudes in the 
tropics and especially down near timber-line, the ' mountain grass- 
land ' is often very luxuriant and inclusive of woody associates 
which do not, however, normally reach a greater height than the 
herbaceous cover, though in some places a tall scrub may occur 
between the Krummholz and fell-fields. 

In some temperate and tropical regions, especially of an arid 
nature, and in keeping with the general reduction of precipitation 
at very high altitudes, dry conditions may prevail above the timber- 
line. Here the effect of aspect is often particularly noticeable. 
Thus in parts of central Asia, steppe-like vegetation may persist on 
the southern slopes of mountains, while, on the shadier northern 
slopes, more luxuriant alpine meadows often occur at similar altitudes. 
In these meadows may be tall forbs, and on the slopes below are 
often many trees — in marked contrast to the arid steppe-like com- 
munities on the south-facing side (cf. Fig. 83, B). Elsewhere, xeric 
grasslands in which the tussocks of the dominants fail to coalesce 
may be common on slopes around 3,000-4,000 metres, below the 
fell-fields, while on plateaux at similar or even higher altitudes, such 
as the ' punas ' of South America and the ' pamirs ' of Tibet, ex- 
tremely severe drought and temperature conditions may prevail and 
the vegetation consist of sparsely scattered tufts of Grasses and 
hardy cushion-plants (Fig. 138). In the high Andes of Peru, etc., 




the cushions formed by some species may be. so large as to resemble 
1 recumbent elephants ' (C. A. W. Sandeman in litt.). 

At the very highest altitudes, for example above 6,000 metres 
(19,685 feet), Lichens are the chief or only macroscopic plants to 
persist, their growth being usually poor. These very high altitudes 
also introduce the so-called ' cold deserts ' of temperate and even 
tropical regions which, surrounded by tundra and other high-alpine 
zones, are developed in the Rocky Mountains of North America 
and the Andes of South America, in the Himalayas (O. Polunin 
voce), and to some extent in the Norwegian Alps as well as on 







r y -&#;* - ^*"J88? 

Fig. 138. — Alpine puna-like formation near mountain summit in Colombia. 
Farther south in South America the cushion-plants are often much larger. (Phot. 

R. E. Schultes.) 

mountain ranges elsewhere. They are often snow- or ice-bound, 
as are of course the ice-sheets of the polar regions, of which the 
most extensive are the Greenland ice-cap in the northern hemisphere 
and the Antarctic ice-cap in the southern hemisphere. Even around 
21,000 feet there may be occasional flowering plants persisting in 
the Himalayas, although crustaceous Lichens are more frequent and 
go higher (O. Polunin voce, and cf. above). 

Whereas in some cases particular zones of vegetation may encircle 
a mountain at a fairly uniform level and preserving a fairly uniform 
breadth, frequently they are tilted, being often higher on the 


equatorial side ; or else they may be narrow on one side, or dis- 
continuous, or merely partial. Moreover as they usually merge only 
gradually into one another, such vegetational zones are often difficult 
to distinguish unless they are in full development. 

Antarctic Types 

The vegetation types of antarctic and adjacent, regions have become 
sufficiently known in recent years for the broad generalization to 
be made that they are reasonably comparable with those of the 
Arctic, even if there is a tendency towards more tussock formation 
and less woody plants in the South. They also tend to be treeless 
to much lower latitudes ; in conformity with our delimitation of 
the Arctic, these treeless regions and their vegetation will be con- 
sidered here as more or less antarctic, even though many of the 
islands lie far from the Antarctic Continent and are commonly 
referred to as subantarctic. Although the floristic composition is 
largely different, the antarctic flora being in general very limited 
and widely peculiar, with often a high degree of endemism, plant" 
communities in low-antarctic regions often look much like those 
developed in the Arctic under comparable circumstances in similar 
situations. Furthermore, much the same range of vegetation-types 
is found in the Antarctic and Subantarctic as in the Arctic, whose 
plant communities we have described and illustrated sufficiently 
above. Consequently a very general account of the main antarctic 
and subantarctic types should suffice. In this it should be recalled 
that antarctic and subantarctic lands are for the most part widely 
scattered, often being extremely isolated and having extraordinarily 
limited floras, which in the case of the ice-free portions of the 
Antarctic Continent consist almost entirely of lowly cryptogams. 

The Antarctic is a very bleak and stormy region possessed of two 
main climatic areas. The northern one may be described as sub- 
antarctic and is maritime, with warmer winters and cooler summeis 
than much of the Arctic, and powerful winds throughout the year. 
There is heavy precipitation in the form of snow or rain, and little 
distinction between the seasons. The subantarctic islands have this 
climate. The other, southern and central region is extremely cold 
and stormy, having a continental climate without summer warmth. 
Thus on the entire Antarctic Continent there is scarcely any place 
having the mean of the warmest month of the year above o° C. 

In contrast to the Arctic, which apart from ice-caps is largely 



free from snow in summer and widely favourable for plant life, the 
Antarctic Continent is permanently ice-covered except for very 
limited areas chiefly about its borders. Here the ice and snow may 
disappear for a brief period in the most favourable season and allow 
some depauperated, almost entirely cryptogamic, vegetation to grow. 
A few ' oases ' up to a reputed 300 or so square miles in area have 
been reported in recent decades to occur at varying distances inland, 
though mainly near the coast. They are largely free from snow 

Fig. 139. — Crustaceous and foliose Lichens on rocks near shore, Goudier Islet, 
Antarctica. From these rock faces, especially where attacked by Lichens, and 
from marine and wind-borne debris, there may accumulate in crevices and depres- 
sions a fair soil on which fruticose Lichens and Mosses often grow. (Phot. 
I. M. Lamb, courtesy of Falkland Islands Scientific Bureau.) 

in the favourable season and are reported to have lakes green with 
Algae but to be otherwise devoid of evident life, though apparently 
they have not been scientifically investigated. Many rock faces 
even well away from the ice appear entirely barren, at least from a 
distance, though in some places especially near the shore a fair 
growth of crustaceous Lichens may be found ; here there may be an 
accumulation of soil, and, growing on it, fruticose Lichens and Mosses 
(Fig. 139). This poverty of the antarctic vegetation especially on 
the Continent must be related to the very unfavourable climate — 


particularly to the coolness of the ' warm ' season and to the frequency 
and persistent strength of the winds — and also to the isolated 
situation which makes immigration extremely difficult. Almost the 
entire Continent belongs to the zone of ' perpetual frost ', only a 
few peripheral areas having a ' tundra climate ', though this is 
enjoyed by most of the subantarctic islands. 

Only two species of vascular plants, a Grass and a Caryophyll, are 
at all well known from the Antarctic Continent (a segregate of the 
Grass is claimed by some to constitute a third species), but there 
are fairly numerous Lichens, Mosses, and Algae, many of which are 
widespread. The microthermic vascular flora of surrounding 
islands, however, includes a fair number of more or less circumpolar 
species. Characteristic components of this flora are Colobanthus 
crassifolius and Lyallia spp. (the former being known from the 
Antarctic Continent, and both belonging to the Caryophyllaceae), 
Pringlea antiscorbutica (the Kerguelen Cabbage, belonging to the 
Cruciferae), Acaena spp. (Rosaceae), Azorella spp. (Umbelliferae), 
and the Grass Deschampsia antarctica (also found on the Antarctic 
Continent). Mixed with various hardy cryptogams, these and other 
vascular plants form a thin tundra which, with increasing luxuriance, 
extends northwards over the Antarctic Islands and into much lower 
latitudes — for example in southern South America. Nevertheless 
this ' subantarctic ' tundra occupies only a tiny area when compared 
with its arctic counterpart. Like the latter, it includes some 
shrubby plants, mostly of cushion-form, and often tussocky Grasses, 
as on South Georgia. 

The Antarctic Continent is largely covered by the world's greatest 
ice-cap and consequently is a vast polar waste. Only here and there, 
on ice- and snow-free spots of the shore or inland ' oases ', on steep 
walls of rock or stony slopes, and on mountain-peaks protruding 
from the ice, are found the Lichens, Mosses, and Algae mentioned 
above — the vegetation at the best being in general far poorer than 
is to be found in all but the most barren of arctic habitats. Even 
Bacteria appear to be relatively few in number. Most of them, and 
many of the largest known patches of macroscopic vegetation, have 
been found in the Graham Land sector of western Antarctica below 
the 68th parallel of S. latitude. Here, in the most suitable situations 
within areas of favoured climate, may be found the two or three truly 
antarctic vascular plant species and patches of cryptogamic vegeta- 
tion that are locally more or less closed — particularly with such 
Mosses as Brachythecium antarcticum and species of Grimmia and 


Andreaea. The most luxuriant vegetation is often found in areas that 
are manured (but not much trampled) by Penguins, etc. Lichens 
inhabit the rocks, and include species of Caloplaca that introduce 
bright colours much as in the Arctic, while Mosses may form an 
almost continuous investment very locally (Fig. 140). But in general 
the vegetation even around the periphery of the Continent is sparse 
and only encountered in occasional favoured areas, most tracts 
being ice-covered and devoid of evident plant growth, though the 

Fig. 140. — Luxuriant growth of Mosses, broken chiefly by rocks bearing Lichens, 
extending up snow-melt gully in area frequented by Penguins, Deception Island, 
Antarctica. Rarely is more luxuriant vegetation to be seen on or near the Antarctic 
Continent. Note the small tussocks formed by the chief Moss, and the more 
typical barrenness of the ground behind. (Phot. I. M. Lamb, courtesy of Falkland 
Islands Scientific Bureau.) 

surface snow when tested has usually proved to contain some viable 
Bacteria and often also spores of Moulds and Yeasts which had 
presumably been carried thither by air currents. 

Further south, in the perpetually frozen zone, only a few sheer 
walls of rock or peaks or other situations that become bare of snow 
in the brief ' summer ' carry a sparse vegetation consisting entirely 
of hardy cryptogams. However, some Mosses and Algae and fairly 
numerous Lichens persist to at least 78 S., sometimes practically 
covering suitable manured areas near the coastal shelf-ice in deep 


bays, while, inland, three Lichens have been reported from snow- 
free rocks of the Queen Maud Mountains within 237 nautical miles 
of the Geographical South Pole. Samples of snow collected here- 
abouts yielded seven different species of Bacteria ; but in general 
we can describe the interior of Antarctica as almost devoid of 
macroscopic vegetation, and supporting precious little microscopic 
growth or even life. 

In the wide seas that surround the Antarctic Continent, there are 
scattered islands and archipelagos whose low summer temperatures 
and vegetational characteristics indicate polar affinities. They 
include Kerguelen Island in about lat. 49 S. and long. 70 E., whose 
barren appearance is widely attributed to very stormy drying winds 
coupled with the low temperature of the ground. The vegetation 
is largely dominated by Azorella selago and to a lesser extent by 
Acaena adscendens, the former often determining the general appear- 
ance of the landscape in sheltered situations in the interior. Thus 
in otherwise desert-like areas it may form tussocks up to | a metre 
high and 1 metre wide. In some places an almost continuous, 
swarded tundra of these and other plants may be developed, and some 
slopes may be more or less green, as may be damp depressions. The 
prevailing winds being westerly, it is chiefly on the sheltered east- 
facing slopes that the most luxuriant vegetation develops. Here the 
Azorella tussocks have associated Acaena, Agrostis antarctica, and 
Lycopodium saururus, which tend to overgrow them chiefly from 
the eastern side, while blocks of rock may be largely covered with 
Lichens such as Neuropogon spp. — again most luxuriantly on their 
eastern sides. 

In suitable situations on Kerguelen the Azorella or other 
' cushions ' tend to coalesce to form a continuous cover which in 
the most favoured ' oases ' may be replaced by almost pure Acaena. 
Species associated with the Azorella are commonly few, though 
usually some crustaceous Lichens are to be found on the stones, and 
there may occur such dicotyledonous plants as Pringlea antiscorbutica, 
Colobanthus kerguelensis, and Lyallia kerguelensis, and the Grasses 
Agrostis antarctica and Festuca kerguelensis. Like the dominant 
Azorella, most of these plants are of tussocky growth, as are associated 
Mosses such as species of Rhacomitrium and Blindia. Acaena forms 
a more even, if wavy, meadow-like community which from a distance 
may look like a relatively smooth heathland. Commonly associated 
with it are the same Pringlea, Galium antarcticum, Ranunculus 
biternatus, and various Grasses. Apart from Pringlea, which may 


represent a relic from some past age, the' plants in the Azorella 
community are all more or less xeromorphic. Those of the Acaena 
community, on the other hand, are more or less hygrophytic in 
character, lacking any obvious means of protection against the 
mechanical and desiccating effects of the wind. Thus from the 
rampant main axes of this dominant, which form a thin-meshed 
network on the ground, leafy shoots ascend to a height of often 
20-50 cm. (about 8-20 in.). In rocky places various Ferns occur, 
including a Filmy-fern {Hymenophyllum peltatum) and forms of the 
familiar northern Polypody {Polypodium vulgare agg.) and Brittle- 
fern (Cystopteris fragilis 8.1.). Limited to the salty beaches are 
Cotula plamosa and Tillaea moschata. 

The vegetation of the other islands of the Kerguelen Group, such 
as the Crozet and Prince Edward Islands which extend westwards 
in comparable latitudes to 38 E. longitude, and the nearer but more 
southerly McDonald and Heard Islands, does not appear to show 
any important deviations in its general character from that of 
Kerguelen Island itself. However, McDonald and Heard Islands 
tend to be particularly barren, presumably owing to their higher 
latitude (c. 53 ° S.). Of the thirty species of flowering plants known 
from the Kerguelen Group, no fewer than six (20 per cent.) are 
endemic, the Pringlea (Kerguelen Cabbage) being the sole known 
representative of an apparently endemic genus. 

Of the islands lying south of New Zealand, Macquarie Island 
(about 55 S.) is antarctic in character, with few and usuallv dwarf 
woody plants (such as Coprosma repens, though the taller Acaena 
adscendens and an allied species also occur). Wide stretches of the 
hills are taken over by the yellowish tussocky grass Poa foliosa, 
between whose tufts occur here and there larger ones of Stilbocarpa 
polaris and silvery rosettes of Pleurophyllum hookeri as well as two 
species of Acaena. Azorella selago is reported to form large cushions 
on the wind-blown summits of some of these hills and to harbour 
other plants as on Kerguelen. On the rocks of the shore are tussocks 
of Colobanthns muscoides, Tillaea moschata, and a small endemic Grass 
besides the more familiar Festuca erecta. In swampy as in some 
drier places Poa foliosa is typical, and on the beaches Cotula plumosa. 
The nearest land is 650 km. away, and as it is considered unlikely 
that any vascular plants survived the severe Pleistocene glaciation 
on Macquarie Island, it is thought that migrating sea-birds must 
have been the main agents of importation of the thirty-live species 
of vascular plants known to grow on the island — cf. page 114. 


Even though they lack arborescent growth and must be men- 
tioned here, most other ice-free islands of the South that have been 
investigated appear to be scarcely polar in type. Thus the Falkland 
Islands near southern South America and the Antipodes Islands 
near New Zealand support quite large bushes, besides Grasses up 
to 15 metres in height which are apt to grow so closely together as 
to prevent the entry of other plants. The large island of South 
Georgia, however, which lies about 1,200 miles east of Tierra del 
Fuego, is within the zone of pack-ice and has a clearly antarctic 
character. In spite of persisting glaciation and poverty in species, 
the vegetation is relatively luxuriant near the shore and in sheltered 
valleys. Its character is chiefly determined by a few plants, such 
as the tussocky Grass Poa flabellata and the somewhat shrubby 
rosaceous Acaena adscendens, which are often so overwhelmingly 
dominant that other plants play only a minor role. The Poa tufts 
may be quite tall, the height of a Man being commonly approached 
by the long and stiff leaves protruding from tussocks that themselves 
often exceed \ a metre in height, and that are separated by bare spaces 
in which, w T hen the area is sloping, water flows away quickly after 
snow-melt or heavy rain. Such vegetation is largely confined to 
seaside situations. On rocks near the beach a thick turf and sward 
may be formed, especially in manured areas, and sometimes over- 
lying deposits of peat. In some places an unbroken grassy tundra 
may extend to an altitude of 200 or even 300 metres on sheltered 
north-facing slopes (which in the Far South of course tend to be 
the most sunny and favourable). Other such slopes, especially 
where damp, and the banks of brooks, may be covered by the 

Inland areas on South Georgia which have not been so invaded 
and are not too swampy, often support a meadow-like community 
of such Grasses as Festuca erecta, Deschampsia antarctica, and a 
relative of the European Phleum alpinum, often with associated 
Acaena adscendens. Mosses and especially Lichens may here play 
an important role in the consolidation of the vegetation, ' Reindeer- 
moss ' Lichens (including a Cladonia allied to C. rangiferina) and 
members of the lichen family Stictaceae being often abundant, while 
Neuropogon melaxanthus and allied species may practically cover the 
rocks in the higher zones. Intermediate situations frequently sup- 
port mixed cryptogamous carpets, and exposed ones little save 
scattered Lichens. Low-lying swampy areas, on the other hand, 
are typically inhabited by a community dominated by Rostkovia 


magellanica, which gives them a characteristic dark-brownish colour. 
Associated are a few other flowering plants and some Liverworts, 
and many more Mosses. The freshwater aquatic plants include 
a Water-starwort (Callitriche antarctica), a Buttercup {Ranunculus 
biternatus), and several Bryophytes besides a greater number of Algae. 
Vegetation-types of fresh and salt waters, and of snow and ice, 
wherever thev may be developed, will be described in Chapters XV 
and XVI. 

Further Consideration 

The only book devoted to truly arctic vegetation is the present author's 
Botany of the Canadian Eastern Arctic, Part III, Vegetation and Ecology 
(Department of Mines and Resources, Ottawa, Canada, National Museum 
Bulletin No. 104, pp. vii + 304 and map, 1948), which gives illustrated 
descriptions of the vegetation-types recognized in the vast eastern parts 
of arctic Canada. Otherwise the above account has resulted from perusal 
of numerous published papers as well as from personal experience in a 
considerable proportion of the regions involved. Much the same is true 
of the brief consideration given to high-alpine regions, concerning which 
the appropriate parts of most of the general as well as regional works 
cited at the end of Chapter XII may be found helpful if further details 
are desired. As if in tribute to the enterprise of its authors, the Journal 
of Ecology, published regularly since 191 3, is remarkably rich in well- 
illustrated accounts of the vegetation of various arctic and alpine regions, 
including some of the most rigorous and difficult of access. 

All truly arctic vascular plant species recognized to date, including 
those mentioned in the above chapter, are described and illustrated in 
the author's Circumpolar Arctic Flora (Clarendon Press, Oxford, pp. 
xxviii + 514, 1959). 

Chapter XIV 


In the last two chapters we dealt with the main types of land 
vegetation found in temperate and polar and some adjacent or allied 
regions. It now remains for us to complete this survey of vegeta- 
tional types developed on the land of the world with some con- 
sideration of those of tropical and adjacent regions. This considera- 
tion will be a mere brief outline that can scarcely do justice to the 
range of tvpes that includes the most luxuriant, complicated, and 
often changeable vegetation on earth. Nevertheless one hopes it 
may help relate some of these types to those of other regions, and 
at least assist the inhabitant of the latter in his appreciation of what 
is found in the tropics. 

Tropical Rain Forests 

These, especially in equatorial regions, constitute the most 
luxuriant of all vegetation-types. They occur chiefly where soil 
conditions are favourable in moist tropical lowlands and where 
there is scarcely a distinct (or at all events net long and severe) dry 
season. Their chief development is : (a) in the Amazonian region 
of South America, whence they range northwards in the Caribbean 
and Gulf of Mexico regions to nearly the Tropic of Cancer, south- 
wards past the Tropic of Capricorn in Brazil, and westwards to the 
Pacific Ocean coast of Colombia and Ecuador ; (b) about the 
Equator in central and western Africa, extending southwards past 
the Tropic of Capricorn in eastern Africa and Madagascar ; (c) in 
western India and Ceylon ; and (d) in the Malayan region whence 
they range north to the Himalayas, northeast to Indo-China and 
the Philippines, and south and east through much of Indonesia and 
New Guinea to Fiji and adjacent archipelagos of the western 
Pacific, with an intermittent extension in eastern Australia well 
past the Tropic of Capricorn. These are the regions in which 
tropical rain forest appears to be the natural climax under present 



conditions, although in most of them it is tending to diminish 
rapidly in area owing to the activities of Man, and in some consider- 
able tracts has disappeared altogether. Replacement is mainly by 
secondary growth on areas of cultivation. In addition, subtropical 
rain forests (which seem best treated here) occur widely in central 
and southern South America, around the Tropic of Cancer in Central 
and North America and in eastern China, farther north than the 
tropical rain forest in the Himalayan region and farther south in 
East Africa, and also in Hawaii and southeastern Australasia. 

The country occupied by tropical rain forests is usually flat or 
rolling, though they may extend up the lower slopes of mountains 
to an altitude of about 1,000 metres (3,281 feet) or even higher. 
In some areas rain falls almost every afternoon and night practically 
throughout the year, in others there are one or two dry seasons 1 of 
not more than three months each. Often the rain will pour down 
for days or weeks, and everything becomes soaked in a thick grey 
mist. The temperature is relatively high and uniform, the annual 
means being normally around 25-26 C, and the rainfall commonly 
totals between 200 and 400 cm. annually, though in places there 
may be much more. The relative humidity also tends to be high, 
being usually above 80 per cent., though comparatively low values 
may obtain for short periods. Some notion of what the climate 
is like may be obtained from the tropical palm houses of botanical 
gardens. But although the light is dazzling when the sun shines 
on the upper canopy from its midday position high in the sky, 
beneath the commonly three stories of trees a sombre gloom prevails, 
the atmosphere being humid and close. Nevertheless some rays 
may penetrate and sun-flecks prevail — and, it seems, be micro- 
climatically and physiologically important. 

In these tropical rain forests it is chiefly in the tree-canopy that 
animal life flourishes — of innumerable and sometimes gaily-coloured 2 
Insects, Tree-frogs, Lizards and Snakes, Birds, Squirrels, Monkeys, 
and so forth, many of which never touch the ground during their lives 
(I. V. Polunin voce). The component plants may lose their leaves 
individually each year or so ; but there is no regular seasonal change 

1 Professor Paul W. Richards points out (in litt.) that although a dry season cannot 
be adequately defined in terms of months with less than a certain minimum rain- 
fall, dry seasons in these regions may be considered as consisting approximately 
of those months having less than 4 inches (about 10 cm.) of rain. 

2 In general, however, protective coloration is more characteristic of rain-forest 
fauna — particularly with the Insects, which tend to be brown or green and to 
harmonize with their environment (J. A. R. Anderson and I. V. Polunin in litt.). 


affecting the whole vegetation, flowering and fruiting going on all 
the time, though with particular species tending to have their own 
definite seasons in these and some other respects. Thus whereas 
in some species the different individuals may lose their leaves at 
entirely different times, more often there is approximate synchroniza- 
tion of this event between them each year — but not between members 
of different species to nearly such an extent as in most temperate 
forests. The dormant buds are most often small and unprotected, 
but frequently develop after several or many years, so giving rise to 
1 cauliflory ' (the formation of flowers on old * bare ' wood), which 
is particularly common in these regions. 

The main plant components of the tropical rain forest are normally 
the following seven : 

1. The forest trees. These form the main structural component, 
sometimes referred to loosely as the ' roof ' or ' canopy ', which is 
typically made up of three more or less separate strata characterized 
by different types of trees. These ' stories ' or ' layers ', as they 
are also called, are usually ill-defined and indeed seldom easy to 
recognize by casual observation, owing to the fact that species of 
all manner of intermediate heights are commonly present, while 
upgrowing young trees may be of almost any height up to the stratum 
to which their kind belong, and even different component species 
of a stratum often have different heights. In general, however, 
there can be distinguished strata consisting of trees whose crowns 
vary in height about a mean, and commonly there are three such 
strata in tropical rain forests. 

The roof of the forest has usually an irregular profile, the trees 
of the highest (A) stratum being often more or less widely spaced 
and rarely forming a continuous layer to which the term ' canopy ' 
may be applied. The second (B) stratum, or sometimes even the 
third (C), is commonly the highest layer of tree crowns forming a 
continuous mass. The crowns of the B stratum typically extend 
from about 15 to 30 metres in height, while the still shorter trees 
composing the C storey are usually small and slender and have 
narrow tapering crowns commonly 5-15 metres high. Fig. 141 is 
a profile diagram of typical mixed rain forest in British Guiana in 
which the three tree strata are barely recognizable. When, as in 
this case, the two upper strata are much broken, the third is usually 
dense ; but when the upper ones are dense the third is apt to be 
much less well developed — as in Fig. 142. In the former circum- 
stances Palms may be prominent, as in the example shown in 

Fig. 141. — Profile diagram of primary mixed tropical rain forest, Moraballi Creek, 
British Guiana, showing all trees over 46 metres high on astrip about 45 metres 
long and 7-6 metres wide. (After Davis & Richards.) 

M. 30 

Fig. 142. — Profile diagram of climax evergreen forest in Trinidad, British West 

Indies. The community is a consociation of Mora (Mora excelsa, marked IV I) up 

to about 45 metres high. The diagram represents a strip about 65 metres long 

and 76 metres wide. (After Beard.) 




Fig. 143, or there may be many tall shrubs. Mostly, however, 
dicotyledons predominate, the large Palms, Bamboos, and some- 
times Tree-ferns being evident chiefly in disturbed areas. 

The trees of each stratum usually represent numerous different 
species belonging to various families, a considerable proportion of 
the lower ones being young members of species that are dominant 


(or more often co-dominant) above. Indeed one of the striking 
characteristics of most tropical rain forests is the extremely mixed 
dominance, so that a species commonly occurs only from one to 
three times in an acre. Local consociations of single dominance 
may, however, be developed (Fig. 142), though in real tropical rain 
forests this appears to be rare. The leaves of the trees are com- 
monly of medium size, having an area of 2,000-18,000 sq. mm. 
They are usually entire and * leathery ', and dark-green with glossy 
surfaces. Thus they belong to the laurel or large-sclerophyll type, 
being mostly oblong-lanceolate to elliptical in outline, often with 
extended ' drip-tips '. However, the type of leaf varies considerablv 
with the stratum, drip-tips, for example, being scarcely ever found 
on leaves of mature trees of the higher strata. In many tropical 
rain forests, foliage extends almost continuously from the herbs on 
the ground to the tops of the dominants, and although many large 
trees are present, this foliage of one sort or another mostly hides 
their trunks (Fig. 144). In other instances the canopy is exceed- 
ingly dense and there is little development of undergrowth and 
ground-covering, so that the trunks of the trees stand out in the 
gloom as huge columns. Often they show ' plank ' buttresses as 
in Fig. 145, where an intermediate amount of ground-vegetation 
is visible. 

Although it is often contended that competition between the trees 
finds expression chiefly in the struggle towards the light, actually 
the different strata have each their own species. Normally these 
individually reach their own particular level before attaining full 
development and thereafter make no attempt to pass that level. 
The arboreal species thus seem to fall into groups having a particular 
height-limit and degree of tolerance to shading by the next stratum 
above — or, in the case of the topmost stratum, presumably demand- 
ing full exposure — the struggle towards the light being mainly in 
the immature stages. Nor does there appear to be any intense 
struggle for root-room — a feature that has been confirmed by Dr. 
R. E. Schultes (in Htt.), with the rider that it is unexpected and 
more ought to be made of it. 

It should be recalled finally that in these tropical rain forests, not 
only flowering and fruiting but also the loss and replacement of 
leaves can take place at any time of the year and in fact normally 
does take place at all times (considering the vegetation as a whole). 
Thus in many species the leaves appear to be renewed annually, 
and individual trees devoid of leaves may be observed in the forest 




Fig. 144. — Another scene of tropical rain forest in the Philippines. Note the 
density of the foliage, which often hides the large tree-trunks. 

Fig. 145. — Base of tree-trunk showing exaggeratedly buttressed roots in tropical 

rain forest. 


at any period, though a new crop soon develops on them. In this, 
as was already indicated on pp. 424-5, particular species tend to 
have their own special times — for example, members of the genus 
Hevea in South America commonly lose their leaves at the end of 
the dry season, regularly, each year, just before flowering. 

2. Herbs, etc. Where the tree strata are not too dense and 
sufficient light penetrates, there may be a fair development of green 
ground-vegetation which, like the dominant trees, is independent of 
external support. Such lower vegetation in moist situations tends to 
be largely herbaceous, Ferns and Selaginellas being often prominent, 
whereas on dry ridges it may consist Jargely of woody plants. In 
other cases a shrub stratum (D), consisting mainly of tallish woody 
plants, may be roughly distinguishable, with, below, a ground-layer 
(E) of herbs and tree-seedlings up to 2 metres in height. The 
shrub layer often includes some coarse herbs such as Scitamineae 
(Bananas, Gingers, etc.) which may exceed 5 metres in height. But 
in general, in spite of the prevailingly warm and humid conditions, 
herbs and other lowly plants are little developed on the ground 
owing to the lack of sufficient light. Thus in lowland rain forest 
any luxuriant herbaceous ground-vegetation is found chiefly in 
clearings and by streams and openings where illumination is above 
the average for the level, while in the interior of the forest green 
herbs — apart of course from epiphytes — are found chiefly as widely 
scattered individuals or scarcely at all. On steep slopes, however, 
more light tends to penetrate owing to the angle of the (lateral) 
rays, and herbaceous vegetation is generally more abundant, though 
still the number of herbaceous species is liable to be far smaller than 
that of different trees. Indeed, in contrast to the situation in 
temperate regions, the herbaceous vegetation in tropical rain forests 
is almost always far less various than the arborescent, and, with 
relatively ' open ' conditions, is more apt to form ' families ' of single 
species. The herbs belong to various (if usually few) systematic 
groups but typically include members of the Madder family 
(Rubiaceae) as well as some Grasses and members of the Sedge 
family (Cyperaceae) in addition to Ferns and Selaginellas, and, 
in the Amazonian region, Marantaceae and Melastomaceae. Their 
foliage is usually thin, sometimes variegated, and very variable in 
shape, in contrast to that of the dominant trees. 

3. Climbers. We now come to the first of the three groups of 
plants that, although they are still green, are dependent on external 
mechanical support and afford the main ' forest furnishings ' ; of 


them the climbers or ' vines ' are generally the most important. 
Indeed the woody climbers, also called lianes (or lianas), are apt to 
be so large and numerous as to afford one of the most impressive 
features of the tropical rain forest. They may be thin and wire- 
like or rope-like, or as thick as a man's thigh, vanishing like cables 
into the mass of foliage overhead, or here and there hanging down 
in gigantic loops. Often they are unbranched up to the profusely 
branched crown. Some are said to attain lengths of over 200 metres, 
ascending one tree, then descending to the ground before ascending 
another, and so on. Often they pass from tree to tree and link 
the crowns so firmly that even if a tree is cut through at the base 
it will not fall. 

Lianes are most abundant where the forest has been disturbed, 
or about its margins — as for example along river banks where they 
may completely screen the interior of the forest. In addition to 
the large woody climbers that reach the crowns particularly of the 
B stratum of trees, there are usually some small, mainly herbaceous 
ones (including Ferns) that seldom emerge from the shade of the 
undergrowth. Among the climbers the large lianes comprise, how- 
ever, by far the more numerous synusiae (groups of plants of similar 
life-form, each filling much the same ecological niche and playing 
a similar role, and contributing to a biocoenosis — cf. p. 321). These 
large lianes belong to many different families and genera — chiefly 
of dicotyledons, though Climbing Palms or Rattans are often 
prominent among them. 

The climbers as a whole include twiners which by the revolving 
movement of their growing tips become wound around their sup- 
ports ; also root-climbers and tendril-climbers which have specialized 
sensitive roots and tendrils, respectively, for attachment ; and 
scramblers which lack such abilities or organs but scramble over 
other plants, often being aided passively in their climbing by 
recurved spines or wide branching. Many species use more than 
one method. As the crowns of tropical trees tend to be less branched 
and less leafy than those of temperate trees of similar size, woody 
climbers help to close the canopy and decrease the penetration of 
light. They may also mis-shape the crowns or constrict the stems 
of trees, though such things are more regularly done by stranglers 
{see pp. 435-6). 

4. Epiphytes. These are plants which grow attached to the trunks, 
branches, and even living leaves of the trees, shrubs, and lianes, 
such situations being the only ones available in closed forests for 


species of small stature but having high light-requirements. A few 
of the larger types and many of the small ones grow in the rain- 
forest undergrowth, being supposedly species that are intolerant of 
root-competition or smothering by fallen leaves. All have to put 
up with lack of soil and hence paucity of mineral nutrients, and a 
more or less precarious water-supply, though in this last connection 
we should recall the persistently heavy and often daily rainfall in 
most of their habitats. 

Epiphytes do not ordinarily have any ill-effect upon the supporting 
' host ', and, though constituting a very characteristic element in 
the structure of the forest, play only a minor role in its economy. 
This may even be the case when the epiphytes are so abundant as 
to form an almost continuous investment of tree-trunks, as they 
commonly do in uplands where the tree-canopy is thin or relatively 
simple (cf. Fig. 162). They do, however, play an important part 
in the ecosystem as habitats for animals, and they are further inter- 
esting in showing many remarkable structural adaptations. Their 
number and diversity are great, usually involving a wealth of crypto- 
gams of all lower groups as well as Pteridophytes and flowering 
plants, including some shrubs. Indeed it is the presence of a wide 
range of epiphytes which especially distinguishes the tropical rain 
forest from temperate forest communities, though epiphytes are 
characteristic also of montane and subtropical rain forests, and may 
be even more luxuriantly developed therein. Moreover, different 
species of trees frequently show distinctions in their epiphytic 
floras, supposedly because of different chemical constituents of rain- 
wash as well as for the more obvious reasons of shade or bark- 
texture, etc. 

A few of the more striking types of adaptations of epiphytes should 
be mentioned. Many are constructed so as to collect a substitute 
soil, which is mainly derived from the dead remains of other plants, 
being often assembled by Ants which inhabit the plant's own root 
system that grows into the so-constituted ' vegetable flower-pot ' of 
these ' nest-epiphytes ' (Fig. 146). Others have to be able to 
absorb water rapidly and for this purpose often have spongy ' vela- 
men ' roots ; they also have to be able to conserve the water they 
get, and consequently are often markedly xeromorphic or possessed 
of special reservoirs (e.g. Fig. 147) or storage tissues (in ' tank- 
epiphytes '). 

Three main classes of rain-forest epiphytes may be recognized, 
corresponding to different microhabitats : (a) extreme xerophilous 




Fig. 146. — An epiphytic Fern (Drynaria sp.) which has small humus-gathering 
leaves and larger photosynthetic ones that also produce spores. ( < about rg".) 

Fig. 147. — An epiphytic Bromeliad (Billbergia sp.) with a mass of fibrous roots 
investing the branch of the ' host ' tree. Note that the leaves form urn-shaped 
cups for collecting and holding water, which is absorbed by special hairs on their 

insides. ( X y.) 


epiphytes, living on the topmost branches and twigs of the taller 
trees, such as some Bromeliads and, remarkably enough, Cacti ; 
(b) sun-epiphytes, usually xeromorphic and occurring chiefly in the 
centres of the crowns and along the larger branches of the upper 
tree-stories, and usually comprising the richest of the epiphytic 
synusiae in both species and individuals ; and (c) shade-epiphytes, 
mainly found on the trunks and branches of C-stratum trees, or on 
the stems of the larger lianes. The shade-epiphyte synusia consists 
chiefly of Ferns, and most of its members show no trace of xero- 
morphy. The average vertical ranges of the different synusiae 
depend on the light factor. Thus they tend to be constant within 
any one type of forest but differ in different types, being high in 
forests where the top strata are dense, relatively low in more open 
types of forest, and still lower on isolated trees or the margins of 
clearings or rivers. 

Further adaptations (or anyway beneficial specializations) widely 
exhibited by epiphytes are wind-borne spores (such as those of 
Ferns) or seeds (such as those of Orchids) or fruits, though other 
seeds and fruits are commonly dispersed by animals. Indeed it is 
difficult to conceive of epiphytes being able to maintain themselves 
without some effective means of dispersal of their propagules. 
Some types, often termed hemi-epiphytes, develop long aerial roots 
which reach the ground and so link epiphytes with the next group 
of ' forest furnishings ', the stranglers. 

The epiphytic vegetation of tropical rain forests often includes 
abundant Algae, Lichenes, Musci, and Hepaticae ; indeed, with the 
usual absence of the mossy layer on the forest floor, all the Lichens 
present and almost all of the Algae and Bryophytes are normally 
epiphytic except occasionally in spots where fallen leaves have not 
collected. There are, however, all manner of ' associules ' on stones, 
fallen logs, and so forth, as well as on tree-trunks down to ground- 
level. Otherwise, non-vascular plants contribute widely to the 
classes of sun- and shade-epiphytes and also grow as ' epiphyllae ' 
on (normally living) leaves — in the last instance mainly in the shady 
undergrowth. Actually, the most abundant epiphytes of the shade 
community tend to be Bryophytes, which often carpet the branches 
of shrubs or hang down in the air, while the epiphytes of the sun 
community include also many foliose Lichens, the Bryophytes of this 
synusia tending to be more compact and xeromorphic. The 
epiphyllae are mainly Algae, Lichens, or leafy Liverworts, and are 
found chiefly on the upper surface of rather long-lived evergreen 


leaves of tough consistency. They often have interesting structural 
modifications which appear to assist their adherence to the sub- 
stratum, though normally they do not have any appreciable ill-effect 
on leaves even when they largely cover their surfaces. 

5. Str anglers. These are plants which begin life as epiphytes but 
later send down roots to the soil, becoming independent or nearly 
so, and often killing the tree which originally supported them. 

Fig. 148. — Roots of Strangling Fig on a large tree-trunk. 

They thus form a synusia which is biologically intermediate between 
dependent and independent plants. Most familiar and plentiful in 
both species and individuals are, widely, the Strangling Figs (Ficus 
spp.), which may play a considerable part in the economy as well as 
physiognomy of the rain forest. The seeds commonly germinate 
far up in the forks of tall trees and from the epiphytic bush first 
formed are developed long roots which descend to the ground, those 
nearest the trunk of the supporting tree branching and anastomosing 
until it is encased in a strong network (Fig. 148). After a time 
the original tree usually dies and rots away, leaving the strangler, 



whose crown has meanwhile become large and heavy, as a hollow 
but independent tree in its place (Fig. 149). Species of Clusia, 
forming large crowns but seldom killing their hosts, are often the 
most plentiful stranglers in the South American rain forest. 

6. Saprophytes. These, the plants obtaining their nutriment from 

Fig. 149. — An old specimen of Strangling Fig in which the roots serve as trunks, 
the original ' host ' having disappeared. 

dead organic matter, together with the parasites, comprise the non- 
green, heterotrophic components of the tropical rain-forest vegeta- 
tion. As in temperate woodlands, the vast majority are Fungi and 
Bacteria which aid in organic breakdown — chiefly near the surface 
of the soil. There are, however, in addition usually some small 
associated flowering plants such as certain Orchids and mem- 
bers of the Burmannia family (Burmanniaceae) and the Gentian 
family (Gentianaceae), as well as others of the Triuridaceae and 
Balanophoraceae, which contain little or no chlorophyll and live 
by the same saprophytic means. They are chiefly found in deep 
shade on areas of the forest floor where dead leaves tend to accumulate 




to a greater depth than usual — e.g. in slight hollows or the angles 
between the buttresses of trees — but they may be absent from 
considerable tracts, especially in rain forest with a marked dry 
season. For in general there is very little humus accumulation in 
the tropics, owing to the rapidity of decay and breakdown of 
organic matter. The forest floor normally is barely covered by a 
thin litter of leaves, and commonly shows through in frequent 
bald patches. 

Fig. 150. — Flower and buds of Rafflesia manillana, a true parasite on the roots 

of a Cissus vine. Rafflesia has no regular leaves or stem, and no chlorophyll, the 

flowers growing directly from the roots of the host; an allied species, R. arnoldii, 

has the largest known flowers (about a metre in diameter). 

7. Parasites. Of these there are, apart from Fungi and Bacteria, 
two main synusiae in the tropical rain forest — the root-parasites 
growing on the ground (Fig. 150) and the semi-parasites (often 
termed hemiparasites) growing epiphytically on the trees. The 
former are few and of little importance, comprising two small but 
remarkable families. On the other hand, the epiphytic semi- 
parasites all belong to the Mistletoe family (Loranthaceae — Fig. 
151) and are numerous in species and often abundant, being met 
practically throughout the area of the rain forest. They are woody 
shrubs that in forests of fair density occur on the twigs and branches 




of the taller trees, whereas in open types or areas they may descend 
almost to ground-level. Their vertical distribution thus corresponds 
with that of the autotrophic sun-epiphytes, and seems to be deter- 
mined chiefly by intolerance of shade. 

In subtropical regions where the rainfall is abundant and well- 
distributed, rain forests occur which are similar to the tropical ones 
except for their tendency to be less luxuriant and dominated by 
fewer species, and to include temperate elements but a smaller total 

Fig. 151. — A tropical hemiparasitic Mistletoe, Viscum orientate, the root of which 
forms a single haustorium (ahsorhing organ). 

flora. Often there are fewer lianes and epiphytes as well as, especi- 
ally, trees ; and often the middle (B) tree storey predominates, 
although traces of both the others may be developed. Examples 
are found bordering on the tropical rain forests and elsewhere as 
already indicated above, and include the more luxuriant ' hammocks ' 
of southern Florida. Such special features as plank-buttressing and 
cauliflory, characteristic of the tropical rain forest, become less 
evident or disappear. As temperate regions are approached, these 
subtropical rain forests pass into the poorer but still evergreen, 
warm-temperate ones described in Chapter XII. It should also be 
noted that, in periodically drier areas inhabited by other types of 


vegetation, there often occur, along rivers, ' fringing forests ' which 
are evergreen and otherwise reminiscent of the rain forest (although 
usually less luxuriant). 

In spite of the widespread destruction wrought by Man particularly 
during the past century, it has been computed that one or other 
form of tropical rain forest or ' seasonal ' forest (see below) still 
occupies about half of the total forested area of the world. 

Tropical Forests with a Seasonal Rhythm 

Where marked dry seasons occur in the otherwise humid tropics 
and the dominants are dependent upon seasonal rainfall, the vegeta- 
tion presents a much more varied appearance. Here conditions 
tend to be so ' critical ' that slight differences in the climate or soil 
may introduce marked changes in the plant formations. Actually, 
regions with one or sometimes two pronounced dry seasons of several 
months' duration occupy a much greater area in the tropics than 
those with a constantly humid climate and luxuriant evergreen 
vegetation. Such ' seasonal ' climates of marked extremes are par- 
ticularly characteristic of the interiors of continents. When their 
areas are timbered and tropical or subtropical, they may be classified 
in the following three main groups of descending water-availability : 

1 . Widespread though variable are the monsoon forests or similar 
' seasonal ' forests, developed in regions enjoying abundant rainfall 
during the wet season, but having this alternating with a pronounced 
drought lasting from four to six months or sometimes longer. The 
total amount of rainfall is usually less than in tropical rain forests, 
being commonly between 100 and 200 cm. per annum if we include 
in this category areas dominated by a rather wide range of deciduous 
types ; and there are marked daily and seasonal changes in tempera- 
ture as well as, commonly, strong winds. Often the character of the 
forest has been changed by human interference, to which its main- 
tenance may even be due, for the dominants are apt to be fire- 

Monsoon or allied forests are found in the areas of the true 
monsoons in India, Burma, Indo-China, and southwards to northern 
Australia, as well as on the margins of the tropical rain forest in 
Africa, Madagascar, Indonesia, and Central and South America. 
Their vegetation is not as luxuriant as that of the tropical rain forest, 


though varying in appearance from an impoverished form of the 
latter down to the savanna-woodland type described below. Thus 
in comparison with the rain forest, the monsoon forest tends to be 
more open, with the trees farther apart and no such scramble of 
all plants for light, while in the dry season (see p. 424) most trees 
shed their leaves and the landscape takes on a ' wintry ' appearance. 
The degree of defoliation depends, however, on circumstances — 
particularly on the severity of the season and the proximity of water- 
courses, along which trees tend to retain their foliage throughout 
the year. Some evergreen trees persist except when the dry season 
is particularly rigorous, and indeed there is apt to be relatively poor 
correlation with the dry season, many species tending to produce 
young leaves some time before it is over. This dry season is often, 
moreover, the period of flowering, so that altogether the monsoon 
forests at this time do not present as lifeless an aspect as do temperate 
deciduous forests in winter. They do, however, resemble these 
forests rather than tropical rain forests in that the trees have thick 
bark, exhibit growth-rings in the wood, lack plank-buttresses, and 
are usually not more than 12 to, at the most, 35 metres in height. 
The trunks of the trees tend to be massive and fairly short, the 
crowns being usually round and large, spreading widely from often 
no great height above the ground. The branches are commonlv 
rather stout and gnarled, and the bark is typically fissured or scaly. 
The community consists of three main tiers — the canopy which is 
liable to be much interrupted, the undergrowth which tends to be 
dense but to have small and hard leaves, and the ground or ' field ' 
layer consisting mainly of more or less lowly herbs. The leaves of 
the trees are usually thin but may be larger than those of the tropical 
rain forest, as in the case of Teak (Tectona grandis). They function 
only during the rainy season and so require no particular protection, 
being commonly hygrophilous (possessed of features characteristic of 
inhabitants of humid situations). The climbers are fewer and smaller 
than in the rain forest, being often herbaceous, and vascular 
epiphytes are normally found only in the canopy. Consequently 
the undergrowth is often luxuriant, consisting of shrubby thickets (or 
sometimes of tall tufty Grasses when the forest is immature, though 
this is rather a feature of savanna-woodland). Bulbous and other 
geophytes are also commonly present in considerable numbers and 
flower in the dry season, whereas the shrubs tend to blossom at 
the onset of the monsoon rains before the leaves appear, and many 
herbs follow suit during the rainy season. Although the Teak 


forests may be relatively uniform, a feature of most other types of 
monsoon forests is the variety of tree-species involved, which may 
number forty or fifty in a single tract. In general, however, both 
flora and vegetational luxuriance are markedly poorer than in the 
tropical rain forest, though the most striking difference lies in the 
seasonal nature of the monsoon forest. 

2. Savanna-woodlands or ' parklands ' are often found where the 
rainless period is more prolonged and the annual rainfall less heavy 
than in true closed forest. The trees are mostly widely scattered 
save in favourable situations (such as occur near watercourses), show 
increasing xerophily and resistance to drought as water-availability 
decreases, and are often leafless during the dry season. The 
vegetation is open and park-like, being rich in terrestrial herbs and 
especially Grasses, but very poor in lianes and epiphytes. Bulbous 
and other geophytes are often abundant. The trees are commonly 
stunted to little more than tall shrubs, being usually much less than 
20 metres high and sometimes overtopped by tall Grasses during 
the rainy season. Although normally various, they are most char- 
acteristically and widely members of the Pea family (Leguminosae), 
which frequently dominate alone. Some of the trees may be quite 
lofty ; but usually they are of lowly stature, often with squat stems 
and thick fissured bark, the crowns being commonly flattened 
or umbrella-shaped — allegedly in relation to wind, though this 
presumption has been questioned. Thus Mr. A. C. Hoyle {voce) 
believes that the causal factors are complicated and include high 
insolation at times of limited water-supply. The leaves of the trees 
are usually xeromorphic and their buds well protected, but flowering 
frequently takes place late in the dry season. 

In some places, as for instance in South Africa, it is contended 
that the more open parkland or ' tree-veld ' is successional, the 
scattered Acacias and other trees attracting Birds and Mammals that 
drop seeds around. In this manner a considerable variety of other 
trees, shrubs, and climbers may be brought in and locally oust the 
tall Grasses, so that a patchy type of savanna-woodland develops. 
Even this may not be truly climax but due to edaphic or biotic 
influences (particularly fire), and indeed it may well be that most 
savanna-woodland is a fire climax in which the trees become self- 
selected for fire-resistance. 

In these well-lighted, open types of woodlands a few small lianes 
and epiphytes may occur, though often the latter, particularly, are 


lacking. On the other hand in favourable situations where the 
trees grow close together, there may be a fair array of xerophilous 
epiphytic Bromeliads, Orchids, and Ferns, that seem more properly 
to belong to the rain forest. 

Savanna- woodland of one sort or another is found very widely 
in tropical and subtropical regions including much of Cuba and 
elsewhere in the Caribbean, Brazil and northern Argentina, East and 
central Africa both north and south of the Equator, and occupying 
much of India and China as well as of northern and eastern Australia. 

3. Thorn-woodlands, with similar or allied types called tropical 
thorn-forests, thornwoods, thornbush, caatinga, etc., are usually still 
more xerophilous, being found in areas of still lower rainfall and 
more prolonged drought-period than the often poorly-differentiated 
but usually much more grassy savanna-woodlands. Indeed Grasses 
are often lacking in the drier thorn-woodlands, or tend to be 
segregated into clumps separated by areas of bare soil. The present 
group of vegetation-types are found chiefly where the annual rainfall 
is between 40 and 90 cm. but variable, with the temperature high 
all the year round, ranging from 15 to 35 C. The terms quoted 
are not synonymous in that they are apt to be applied to communities 
of different physiognomy growing under different circumstances in 
different regions. And just as the major types dealt with previously 
in this section grade into one another, as indeed do many vegetation- 
types elsewhere, so do the present variants intergrade and inter- 
digitate with the savannas and grasslands dealt with below. More- 
over, various and even quite different types and sequences may be 
involved where local water or other conditions change markedly. 
Thus succulents are often most in evidence in areas of particularly 
coarse, over-drained soil, and may form distinctive communities 
alternating with one or another type of thorn-woodland. 

The foliage of the dominants of tropical thorn-woodlands is 
deciduous or markedly xerophilous, or often reduced to mere scales, 
thorns or prickles being a common feature as the above names 
imply. Switch-plants with woody photosynthetic stems are also 
characteristic of the community. The roots of the main plants are 
much branched, and competition among them for water may be 
severe. Often they penetrate very deeply, shallow-rooting types 
such as Grasses having little chance of success. Many of the woody 
plants store water for the drv season in swollen trunks or roots, as 
in the case of the Brazilian Bottletree {Cavanillesia arbor ea) whose 


trunk swells to form an almost barrel-shaped structure. There are 
also sometimes tracts dominated by arborescent succulents, including 
giant Spurges (Euphorbia spp.) in the Old World and characteristic 
members of the Cactus family (Cactaceae) in the New World. A 
few xeromorphic herbs may occur, such as terrestrial Bromeliads 
with sharp-edged leaves. The accompanying shrubs are no less 
xeromorphic, being often thorny Acacias or other members of the 
Pea family, and sometimes forming a grey bushy jungle 3-5 metres 
high, while a few thin woody climbers may also occur. Epiphytes 
are usually absent, though Spanish-moss (Tillandsia sp.) may be 
plentiful locally. With the onset of the rainy season the leaves and 
flowers emerge and vast numbers of geophytic herbs may spring 
from the soil, the rhythm of life being even more strongly marked 
than in the monsoon forests. 

Tropical thorn- woodlands and -parklands, etc., are widely developed 
in dry regions : for example, in northeastern Brazil and elsewhere 
in tropical South America, in the islands of the Caribbean as well 
as in Central America and Mexico, and in and about the Sudan and 
regions bordering on the southern Red Sea and the Gulf of Aden. 
They also occur in southwestern Africa, in India, and in the west 
and centre and northern hinterland of Australia. Thorn- woodlands 
and allied types often occupy sandy or limestone soils that are very 
permeable to water, and alternate with savanna which tends to 
prevail on the stiffer soils that retain rainwater near the surface. 
In places where there is a local increase in humidity, as for example 
in depressions or ravines, this savanna passes to savanna-woodland. 

Tropical and Subtropical Savannas and Other Grasslands 

As in other major regions, grasslands in the tropics and subtropics 
constitute one of the main vegetational types and are usually, but 
not always, dominated by members of the Grass family (Gramineae). 
For sometimes grass-like plants of other affinity (particularly of the 
Sedge family, Cyperaceae), may be the dominants over considerable 
areas, or plants of quite different types may play a similar role. 
Savannas are by some considered synonymous with treeless grass- 
lands or steppes but in the present work the term is employed to 
denote areas which ecologically speaking appear to be true grass- 
lands, in that Grasses or similar herbs seem to be the real dominants, 
but in which trees or tall bushes (in ' bush savanna ') occur in open 
formation and give a particular character to the landscape. Nor- 


mally these tall woody plants are more or less widely scattered except 
in unusually favourable circumstances such as obtain along water- 
courses or elsewhere that water is relatively plentiful. The occur- 
rence of more hygrophilous ' meadows ' is rare in the tropics and 
is clearly due to some particular local factor or factors of disturbance. 

Although many grasslands in the tropics, as elsewhere, appear to 
be ' natural ' to the extent that they do not owe their existence to 
direct interference by Man, it now seems clear that there is no such 
thing as a ' tropical grassland climate ' and quite possible that 
tropical grasslands are not in fact climatic. This may even be the 
case with the savannas which in one form or another constitute 
their most common expression, for in the tropics taller woody plants 
are rarely absent from such areas. Certainly many of these tracts 
owe their persistence or very existence to fires or browsing animals, 
or are edaphic climaxes due to local soil conditions, while others 
appear to be serai. Thus fires often destroy woody and other 
dicotyledonous plants and Palms without appreciably damaging the 
underground parts of the Grasses. But whatever their ecological 
significance may be, these grasslands constitute characteristic types 
of considerable economic as well as areal importance. 

Savannas or their treeless counterparts are very widespread in 
tropical and subtropical regions, where they often cover vast tracts— 
though not without considerable local variation within their own 
areas, as in the cases of many South American ' campos ' and 
' llanos '. Examples are to be seen in southwestern North America 
and the West Indies, in Central and South America both north 
and south of the Amazonian forests, and in very many parts of 
Africa such as the Sudan and within as well as around the closed 
forests of the Congo, etc. They also occur in central Madagascar, 
in disturbed and upland areas of India and elsewhere in Asia, and 
to the north of the central desert tracts of Australia. The climate 
is hot, with a moderate range of temperature and a fair rainfall 
often exceeding ioo cm. annually and well spread over 120 to 190 
days, during which ' rainy season ' large areas may be constantly 
under water. On the other hand there is a prolonged drought 
lasting often for six or seven months of the year, and a tendency to 
desiccating winds. With such substantial rainfall, more or less 
xerophilous woodlands are apt to predominate in the absence of 
disturbance — at least elsewhere in areas of very high temperatures 
and prolonged rainless periods during the vegetative season. These 
woodlands may include the savanna-woodlands, which are dis- 


tinguished by the fact that in them the trees appear to be dominant, 
although such types grade into savannas, even as these last grade 
into treeless grasslands. 

The savanna presents mostly a park- or orchard-like appearance 
— a landscape typically of plains of tall Grasses with scattered trees 
and shrubs. In hollows or swales the trees frequently grow close 
enough together to form woods, whereas on ridges they are sparse 
or wholly absent, the vegetation thus constituting a steppe. The 
Grasses commonly exceed the height of a Man, but range, in different 
instances, from less than i to more than 4 metres high, and form 

Fig. 152. — Palm-savanna in southern Florida. 

a yellowish straw frequently topped by silvery ' spikes \ They 
typically grow clustered in dense tufts which exhibit, in the intervals, 
patches of bare soil often of a reddish or yellowish hue. Low 
bushes with small and hard evergreen leaves and often prickles or 
thorns may occur among the Grasses. 

The trees which appear at greater or lesser intervals in the savanna 
are usually stunted and gnarled but sometimes lofty. While many 
are deciduous, others are evergreen ; common heights are 3-6 
metres. They belong to characteristic species not usually occurring 
in the forest, and commonly include Palms (Fig. 152) or other 
plants of peculiar habit (Fig. 153). Often the fast-growing, coarse 
and stiff Grasses interpenetrate the lower branches of the trees and 
remind the ecologist of their tendency to dominate. Elephant Grass 
(Pennisetum purpureum) may exceed 5 metres in height and form 
an almost impenetrable ' thicket '. The trees typically include some 




Acacias and other members of the Pea family (Leguminosae), and, 
in Africa, the Baobab (Adansonia digitata), with its hugely swollen, 
water-storing trunk. They often have thick and corky, fissured bark 
and in favourable situations may form groves. In uplands the 
Grasses tend to be lower and more mixed with forbs, though even 
in the lowlands some forbs are to be found — both hardy perennials 
and others having tubers or bulbs, which enable them to burst 
forth into leaf and flower with the recurrence of the rainy season. 

Fig. 153. — Savanna in Australia under rainfall of 2=5-75 cm - annually. 
(Phot. D. A. Herbert.) 

Although many present-day savannas and treeless grasslands have 
probably resulted from clearance of closed forest, they usually 
experience a longer period of drought each year than do forests. 
However, they have more frequent rains and usually less permeable 
soil than the relatively arid thorn-woodlands. Thus the rainwater 
that falls is readily available for the shallow-rooting, tussocky 
Grasses whose dead straw also accumulates to form a mulch, and 
whose close ' felt ' of roots further helps in water-retention. On 
the other hand, tree-growth is largely restricted to places where 
the ground-water lies at no great distance from the surface, most 
of the successful trees having roots penetrating deeply enough to 


tap this more lasting source. They also tend to have relatively low 
and compact crowns, often of spreading, umbrella-like shape, 
allegedly as a result of exposure to winds. But in many areas they 
are kept at bay or ousted by burning, or by the intensive grazing 
of herds of wild or domesticated Mammals ; of these, many of the 
world's largest are inhabitants of the great grassy plains of tropical 
and subtropical regions. 

Semi-Desert Scrubs 

The arid bushlands characterized by scrubby Acacias and other 
xeromorphic shrubs are often included among deserts, though 
actually it seems preferable to consider them as transitional between 
true deserts and savannas or thorn-woodlands. In tropical and sub- 
tropical regions they are found particularly on stony or rocky 
hill-sides, in open rolling country, and on sandy or gravelly areas 
exposed to the full glare of the sun. Examples may be seen in 
the southwestern United States and adjacent Mexico, along the 
foot of the Andes, and in Africa especially bordering on the Sahara. 
In the East they occur in Arabia and elsewhere near the northern 
shores of the Indian Ocean, and also in Australia. The climatic 
conditions supporting these warm-region scrubs lie between those 
of the desert proper and of thorn- woodlands, the temperatures being 
variable but high at all seasons, and the rainfalls occasional though 
seasonal and commonly averaging from 20 to 50 or more cm. 
annually. Important is the seasonal distribution and general 
reliability of the rainfall. 

Many of the plants are veritable caricatures, such as those that 
are grotesquely swollen to store water. 

The bushes in these semi-deserts are often of fair size and either 
grow separated or aggregated into a more or less continuous scrub, 
most often of thorny Acacias. Grasses if present are dwarfed and 
wiry, and usually reduced to isolated bunches or ragged patches. 
Other herbs tend to be leathery- or fleshy-leafed, or, in numerous 
instances, geophytic, with underground storage of food and water. 
Cacti in the New World and cactus-like Euphorbias in the Old 
World frequently form a characteristic feature of the usually open 
vegetation. Like some of the other plants, including the thorny 
Acacias, they are often beset with prickly spines. Large-leafed 
Agaves, Aloes, and Yuccas are also characteristic inhabitants, as are 
smaller succulents. Although some of the shrubs may be fairly 


dense and exceed the height of a Man, they' are rarely close enough 
together in these semi-desert bush-lands to obscure the entire horizon. 
Fig. 154 shows a relatively well-vegetated area in Australia. 

In some subtropical regions such as occur in southwestern North 
America, a shrubby climax may be developed where the annual 
rainfall averages as little as about 8 cm., provided it is reasonably 
distributed and reliable. The dominants tend to be many-stemmed 
but sparsely open, with widely-spreading roots. Although the 

Fig. 154. — Semi-desert ' bush-land ' in Australia. 

formation is commonly called ' desert-scrub ' it seems best men- 
tioned here, even as its more temperate counterpart, characterized 
by the Creosote-bush, was described under semi-deserts in Chapter 
XII. Indeed the southern, subtropical extensions towards Central 
America have little of a regular nature to distinguish them from the 
northern type, and much the same is true in other regions. Thus 
while the dominants tend to be lower, the Creosote-bush being in 
the South only about a metre high and others (such as Bur-sages, 
Franseria spp.) commonly lower, the transition is normally so 
gradual that no further type emerges or account need be given 

14] vegetational types of tropical lands 449 

Tropical and Subtropical Deserts 

The hot deserts are areas in the tropics and subtropics having 
such very slight precipitation that they typically support at best 
only a scanty growth of scattered plants. Even if the nights are 
cool, with dew and mists occurring especially in central regions, 
the days are normally blazing hot. The environment is thus severe 
and the general run of habitats so unfavourable that they can only 
be colonized by appropriately adapted plants and animals. Bald 
expanses of flat or rolling plains predominate, all sunbaked and 
windswept, with wide tracts of yellowish sand-dunes or browner 
gravels or more rugged rocky floors, and in places broken scarps 
of naked hills. Sparingly dotted in the less inimical situations may 
be low and dry, strange-looking plants, or, in the very occasional 
damper situations, luxuriant but limited oases. 

Hot deserts chiefly occur well to the north and south of the 
equatorial zone, which imaginative generalizers are apt to say is 
' hemmed in ' by them. Although probably more extensive than 
they used to be, owing to interference by Man and his domestic 
animals, they are often not nearly so vast and invariable as is popularly 
supposed, great areas being occupied by other types which are only 
relatively speaking ' desertic '. The main examples are the Sahara 
and Arabian Deserts, occupying, respectively, much of northern 
Africa and southwestern Asia, whence there are extensions eastwards 
into northwestern India and northwards and then eastwards into 
temperate central Asia. Extensive hot desert and near-desert 1 areas 
also occur in central Australia and the southwestern portions of 
North America, and smaller ones in southwestern Africa and western 
South America. 

The primary cause of hot deserts is paucity or even perennial 
absence of rain, though an excessively clear and dry atmosphere, 
with scorching sun, usually contributes to the general aridity. 
Thus the relative humidity in the daytime is commonly less than 
50 per cent, and may drop as low as 5 per cent. The rainfall usually 
averages less than 20 cm. per annum, often very much less, while 

1 To many who have experienced the real deserts of, for example, northern 
Africa and southwestern Asia, it seems unreasonable to refer to the more pro- 
ductive of the so-called deserts of North America as properly desertic — hence 
the use of this designation here. It was one of the present author's earliest 
experiences in Iraq to show pictures of such North American ' desert ' areas to 
students who asked ' But, Professor, how can you call those areas desert, when 
the vegetation is so great ? ' 


in some areas there may be no rain at all for several years on end. 
Such areas may be wholly devoid of macroscopic plants over some 
tracts. Intense radiation and considerable changes of temperature 
are also common, so that where the ground is rocky or clayey it is 
liable to be much fissured. Often it is gravelly, sandy, loamy, or 
stony — but none the less arid. Thus although the substratum and 
even the topography may change greatly in a single desert area, 
the abiding influence is the paucity of water. Tracts of different 
type may exhibit different forms and degrees of vegetative develop- 
ment, or sometimes virtually none, but all have the desert character, 
the stamp of aridity. For example in the western Sahara there are 
the pebbly-clayey areas with cushion-plants and succulents, the 
sandy or gravelly beds of dry watercourses populated with Tamarisks 
(Tamarix spp.), the sand-dunes dotted with heath-like bushes and 
grass-tussocks, and the rocky plateaux, most desolate of all, consisting 
of split stones and broken rocks with an occasional spiny or other 
xerophyte anchored in the fissures. In addition there are saline 
depressions which at best support relatively sparse colonies of 
dwarfed shrubby halophytes. 

Similar ranges of type are found in other deserts, or extremes from 
absolutely bare moving sand-dunes to fairly dense heathlands char- 
acterized by switch-plants. There may even be open miniature 
woodlands, with or without large bushy succulents. The American 
near-deserts 1 are remarkable for their giant Cacti such as the Saguaro 
or Sahuaro (Carnegiea gigantea) and their small pebble-like Pin- 
cushion Cacti {Mammillaria spp.), as well as for their glutinous 
Creosote-bushes (Larrea spp.) and characteristic Ocotillo (Fouquiera 
splendens) (Fig. 155). The South African desert is famous for the 
unique gymnospermous Tumboa {Welwitschia mirabilis) and the 
Desert Melon (Acanthosicyos horrida), as is the Australian for other 
highly peculiar plant forms. Thus whereas the desert populations 
are normally limited to relatively small and scattered plants, many 
of which are thorny, each area may have its own particular character 
given by the plants themselves. 

Desert plants are adapted in various ways to withstand the adverse 
conditions under which they have to establish themselves, grow, 
and ultimately reproduce. Many, particularly among the shrubs 
and more occasional shrubby trees, have long roots that are said to 
reach down sometimes to a depth of 10 or more metres (they may 
certainly exceed 15 metres in length) to subterranean water or at 

1 See footnote on preceding page. 



45 * 

least to damp layers deep down in the ground. Regarding the 
4 extraordinarily deep-penetrating root systems ' of Tamarisks, it is 
even reported that they ' could be followed during the building of 
the Suez Canal in places to a depth of 50 metres ' (transl.). 1 Other 
desert etc. plants, especially among cryptogams, endure drought by 
drving up almost entirely without harm to themselves. Yet others 
are denselv tufted or compacted, and often in addition closely 

Fig. 155. — Arizona near-desert scene showing the giant Saguaro (or Sahuaro) 
Cactus {Carnegiea gigantea) and bushy Ocotillo (Foiiquiera splendens). (Phot. 

F. Shreve.) 

invested with hairs and spines, while many, such as Cacti and cactus- 
like Euphorbias, store water in their massive stems or other swollen 
organs. Usually these ' succulents ' have an extensive system of 
roots spread out near the surface of the soil and ready to absorb 
considerable quantities of water when it comes, for most deserts have 
a short rainy season during which conditions are fairly favourable 
for plant growth — especially with the aid of the rich nocturnal dew 
which may occur. At such times annuals spring up and quickly pass 
through their whole cycle of development, 2 while geophytes, with 

1 K. Rubner, Neudammer forstliches Lehrbuch (Neumann, Berlin, 1 Lieferung, 
p. 180, 1948). 

2 After the heavy rains in central Iraq in the spring of 1957, the author observed 
small Plantains {Plantago spp.) and Grasses (especially of the genus Schismus) and 


underground bulbous or tuberous storage organs, send up aerial 
shoots which flower and fruit but die down with the resumption of 
drought. Hence the ' flowery carpets ' of delicate mesophytes that 
some of the more seasonally-varying deserts often exhibit between 
their scattered bushes after adequate rainfall. In contrast to these 
ephemerals and ' deciduous perennials ', the shrubs commonly have 
very small evergreen xerophilous leaves, or sometimes larger 
deciduous ones which are lost after the rainy period ; others may 
have the leaves reduced to scales, photosynthesis being carried on 
instead by green twigs or leaf-like or succulent stems. 

Most of the desert plants having perennial aerial parts are extremely 
xeromorphic, exhibiting such features as excessive development of 
fibrous tissues, thickened or otherwise covered epidermis, sunken 
and protected stomata, and reduction or ' waxing ' of the transpiring 
surface. They also commonly exhibit the xerophytic feature of high 
osmotic value of the cell-sap. Frequently several of these char- 
acteristics are shown by a single plant, sometimes to an extraordinary 
degree. Dispersal of seeds from whole plants detached by the wind 
and acting as tumble-weeds is fairly common, and often seeds will 
remain dormant for years on end before germinating when sufficient 
water becomes available. Even if many xeromorphic desert plants 
may transpire fairly rapidly when water is plentiful, they are able 
when necessary, by such features as those already mentioned and 
by keeping the stomata closed all day, to reduce their water-loss to 
a minimum and so often survive prolonged drought. They also 
exhibit considerable resistance to wilting and to injury as a result 
of water-loss. Thorny shrubs or broom-like switch-plants and 
other drought-endurers are particularly characteristic of deserts, as 
are, of course, extreme succulents and many ephemerals, but it is 
too ' convenient ' to classify desert plants into any such stereotyped 
categories. A few root-parasites also occur rather widely in deserts. 
Contrary to popular supposition, the larger succulents are unable 
to withstand the conditions of the drier deserts. 

Where there is a lasting supply of water, as along the banks of 
rivers whether permanent or seasonal, or where the ground-water 
rises to near or sometimes above the surface, as in oases, the vegeta- 
tion is able to demonstrate at once the natural fertility of the soil 

other ephemerals in his desert quadrats to grow up, flower, ripen seed, and die 
down— all within a period of about five weeks, though this is only in the warm- 
temperate belt. It is, however, possible that germination had taken place before 
the main rains came, and so closer observations must be made in future. Fig. 156 
shows the ' before and after ' effect of heavy rain in a desert of central Iraq. 



>~. ,, ,v 


Fig. 156. — Areas of desert in central Iraq. A, metre quadrat showing only 1 
small perennial tuft (on right-hand side, about two-thirds of the way back), although 
numerous tiny seedlings are appearing after heavy rain. B, close-up of an adjacent, 
less gravelly area a very few weeks later, showing species of Schismus (slender 
Grass), Malva (broad dark leaves), and Plantago (thin rosettes) developed to 
maturity. The nails projecting from the frame are 10 cm. apart in both A and B. 


— provided, of course, that salts are not present to excess. Date 
Palms {Phoenix dactylifera) and attractive gardens can thus be 
cultivated in otherwise desert areas, and in large oases a fine variety of 
tropical and subtropical agricultural crops are produced. Wherever 
there is feed for Mammals, desert forms such as Gazelles are apt 
to pasture, while for miles around inhabited oases little save poisonous 
or distasteful plant material is normally left undisturbed. Along 
dried-up watercourses (wadis) the trees may attain large dimensions, 
though usually remaining small-leafed and thorny, while perennial 
Grasses, which are otherwise rare, often inhabit the sandy or gravelly 
beds. Where desert conditions extend into temperate regions, as 
in the Gobi, the oases are characterized (as mentioned in Chapter 
XII) by tall Poplars and Willows. Here other vegetation is in 
keeping with the temperate situation, the crops being such temperate 
ones as Barley, Wheat, and Plums. On their poleward side such 
cooler deserts are usually bordered by wide steppes, whereas hot 
deserts are typically bordered by semi-desert scrub, at least on the 
equatorial side. 

Mangrove and Other Sea-Shore Vegetation 

By far the most characteristic and important vegetation-tvpe of 
tropical and subtropical sea-shores is the ' mangrove ' or ' mangrove- 
swamp forest ' developed on mud-flats which are exposed at low 
tide but otherwise normally covered by salt or brackish water, at 
least being reached occasionally during the highest tides. Par- 
ticularly favourable conditions for the development of mangroves 
are found in creeks and quiet bays ending river estuaries, where 
tidewaters cause the deposit of river sediment. On the resulting 
flats and deltas, the water-borne seeds or seedlings of the colonizing 
plants grow, soon forming the characteristic, rather low and dense 
forest (Fig. 157). In other cases the mangrove forest in its interior 
may consist of sizeable trees and be quite lofty (Fig. 158). In 
Malaya and to a lesser extent in Borneo there are large stretches of 
uniformly tall, mature mangroves (I. V. Polunin voce). Thus in 
Malaya the ' climax ' mangrove forest consists of relatively few species 
which tend to be gregarious, producing stands of uniform height, 
whereas in Borneo the stands are not so pure or uniform (J. A. R. 
Anderson and I. V. Polunin in lift.). Actually, real climax man- 
grove now scarcely exists in Malaya, owing to felling on a rotation 
of about 30 years, which leads to a retention of this height-uniformity 


Fig. 157. — A typical Mangrove plant, Rhizophora candelaria, forming a char- 
acteristic marginal ' mangrove ' and showing prominent prop-roots below. 

Fig. 158. — Interior of Philippine mangrove-swamp forest at low tide, showing, 

below, the aerating prop-roots and the conical erect aerating roots which project 

upwards from the mud. 


(J. Wyatt-Smith voce). Frequently, however, the interior tallness 
situation is reversed in that the fringe of the mangrove, at least 
where it does not consist of young pioneer plants, is made up of tall 
trees, the interior being of much lower or bush-like plants. This 
is due to the loss of true mangrove conditions and an approach to 
those of the hinterland forest (J. Wyatt-Smith voce). 

The abundant strut-like and often arching prop-roots of the 
mangrove trees or lower dominants, among other features, cause 
deposition of silt and building up of the surface ; often, new mud- 
flats are formed and the forest may extend year by year. These 
prop-roots are well seen in the illustrations, while Fig. 158 show r s 
also numerous slender conical aerating roots growing vertically out 
of the mud. Both types of roots have ' breathing-pores ' and con- 
tain numerous air-spaces that serve for the conduction of oxygen 
to the underground parts of the system. This function is rendered 
vitally important by the nature of the substratum and by the 
usually frequent inundation, and is performed in some cases by 
knee-like or keeled projections of roots above the surface of the mud. 
Another function of the ' pneumatophores ', as the aerating roots 
are called, appears to be to help keep pace with the tendency of the 
surface level to rise through deposition, for their underground parts 
frequently bear the fine rootlets on which the tree is dependent 
for absorption. Different types of mangrove plants bear these 
different kinds of roots ; and the species may be mixed together or, al- 
ternatively, segregated in more or less pure stands. In mangroves 
in general, and particularly in those of the Indo-Malayan region, 
there is often to be found a fairly definite succession. The stages 
of this are usually characterized by different species, and range from 
the pioneers growing on almost continually submerged surfaces to a 
mature mangrove forest of often tall trees whose bases may be 
inundated by only the highest spring tides or, in some instances, 
scarcely ever reached at all. 

Many of the characteristic dominants of mangroves have another 
feature in common, namely, the ' viviparous ' development of the 
seeds — that is, their germination while still within the fruits and 
attached to the parent plant. The typical arrangement, exhibited 
for example by the Red Mangrove (Rhizophora mangle), is for the 
primary root of the seedling to burst through the hanging fruit and, 
with adjacent tissues, to grow down as a long and dart-like, slender 
but bottom-heavy structure (cf. Fig. 26, B). Later on the seedling 
drops — root downwards, so that the tip may be driven into the 


mud * if the tide is out — and forms anchoring lateral roots in a matter 
of hours, often continuing to grow in situ. Actually the seedling is 
buoyant, so that if the tide is in or for some other reason the root 
does not stick sufficiently in the mud when it drops, it may be 
transported by water to some other situation, there to resume normal 
growth if conditions are favourable. Young seedlings are seen 
growing in fair numbers among the roots in the foreground in 
Fig. 158. 

Mangroves often extend some distance inland in brackish swamps 
and lagoons, forming a fairly continuous fringe, or occupying islets 
between which run the sluggish tidal streams. At high tide they 
appear like a flooded forest or, about their low and tangled margins, 
like a mass of green or greyish foliage sitting on the water. Some- 
times they are replaced by Palms (such as Nipa fruticans) or other 
large monocotyledonous plants, while near their climatic limits they 
tend to form dense tall thickets rather than forests. Recession of 
the tide even in the forested types reveals an ungainly mass of muddy 
roots and often grotesque boles. Even the trees are liable to be 
mis-shapen and lowly, while bubbles of stinking gas rise from the 
rotten mire, and a teeming population of crawling creatures adds to 
the atmosphere of gloomy squalor. 

The climate is usually hot and humid, and conditions are kept 
monotonous by the tides and salinity, though there may be alternat- 
ing heavy rainfall and scorching sun. The dominants are evergreen 
and halophytic, the foliage being leathery, fleshy, or protected by 
a glossy exterior or woolly covering against excessive transpiration. 
Their growth is often so manifestly dense as seemingly to prevent 
the entrance of herbaceous or other vascular plants, though indeed 
there are few of these which are adapted to the very specialized 
habitat. A dark coating of Red Algae may occur on the stems and 
roots submerged by the tides, and some Lichens are usually to be 
found on the stems well away from the water ; but other epiphytes 
are rare. 

The chief development of mangroves is in southern and eastern 
Asia, with extensions to northern Australia and the Pacific. Man- 
groves also occur about Central America, and, to the east, with 
little variation on both sides of the Atlantic. In areas of low pre- 
cipitation and general aridity, mangroves and other maritime wood- 
lands are usually lacking or only poorly developed, even as are 

1 According to Dr. Frank E. Egler (voce), this happens far less frequently than 
the text-books imply. 


forests inland. Exceptions may be afforded by the mouths of large 
rivers, where the salt water is diluted with fresh, and conditions of 
physiological drought are thereby relieved — provided cloudiness 
reduces insolation and, consequently, transpiration. 

Although mangroves are usually serai within themselves, and in the 
most favourable of tropical rain-forest areas appear to be succeeded 
by freshwater swamp-forest (see pp. 461-3) or perhaps sometimes 
bv tropical rain forest, in other instances there is no certainty that 
they can develop alone, by mere accumulation of silt or humus, 
into normal land vegetation or even littoral forest. In such instances 
the more advanced and stable ' inland ' types would seem best 
considered as edaphic climaxes, that appear likely to persist in the 
absence of disturbance. 

The other vegetation-types of tropical sea-shores are more or less 
comparable with those of temperate regions. Thus the beach 
between tide-marks is usually devoid of vegetation on sandy or 
shingly shores and bears only Algae on rocky ones, while even above 
high-water mark on exposed coasts the sandy tracts are often poorlv 
vegetated, as are the outermost dunes. However, these last tend 
to be bound by Grasses such as Spinifex Httoreus, whose rhizomes 
give off tufts at intervals, much as do many of the sand-binding 
Grasses of temperate regions. Many other littoral plants of the 
tropics adopt a similar trailing habit — including Pes-caprae (Ipomoea 
pes-caprae), which forms a particularly characteristic and widespread 
vegetation-type. Such plants also have the important faculty of 
being able to grow out of the sand when covered by its drifting, 
and in dry climates their areas of prevalence may extend far inland. 
Some other colonists have prop-roots that grow down and anchor 
them in the shifting sand, and almost all have a very deep and 
extensive root system. In more sheltered situations, shrubs often 
become numerous, as may in time small trees, such as Screw-pines 
(Pandanus spp., Fig. 159, A). Farther back still — or in quiet creeks 
sometimes near high-tide mark — a closed woodland is typically 
formed in areas of sufficient rainfall. In subtropical regions, as for 
instance the extreme southeast of Iraq bordering on the Persian 
Gulf, there may be salt-marshes reminiscent of those of temperate 
estuarine flats. An example is seen in Fig. 159, B, where the planted 
groves of Date Palms (Phoenix dactyUfera) seen on the horizon afford 
a characteristic relief. 

Littoral woodlands developing out of reach of the highest tides, 
for example on sandy and gravelly shores that still retain an abnormal 


-:: m MmS^ 




1-- ^^m:£^'^. 

Fig. 159. — A Screw-pine and a subtropical estuarine salt-marsh. A, a Screw-pine 
(Pandanus tectorius) with prop-roots. Such plants are very common along the 
strand in the eastern tropics and are widely planted for ornamental purposes. 
( X 3V) B, a subtropical salt-marsh in the estuary of the Shatt al-Arab, near 
Fao in the extreme southeast of Iraq bordering on the Persian Gulf, showing, 
behind, a characteristic grove of planted Date Palms (Phoenix dactylifera.) 



amount of salt, are inclined to be highly characteristic and only 
gradually, in space or in time, able to take on the aspect of the local 
hinterland climax. Often they develop as a belt just inland of the 
mangrove. In other places the forest or scrub even in the tract 
lying nearest to the sea may be devoid of halophytic species. Where 
salinity prevails and the littoral forest differs from that of the general 
hinterland, the sandy or stony soil is often almost bare of dead leaves, 
and the trunks of the trees are commonly naked ; or they may be 
beset with epiphytes, both thick-leafed and cryptogamic, and support 
a mass of thin-stemmed climbers. Where the trees are less close 
together, there is often a dense undergrowth of shrubs and small 
trees, or patches of coarse Grass. The leaves are usually leathery 
or succulent, often hairy when young, or hard and sword-like in 
the cases of Screw-pines and the leaf-segments of Coconut (Cocos 
micifera) or other Palms. Particularly characteristic trees in such 
situations in the Old World are species of Barringtonia. As the 
distance from the coast increases, protective measures become less 
necessary and pronounced, and the forest takes on more and more 
the appearance and flora of the local climax, containing few r er and 
fewer species which are not to be found away from the influence 
of the sea. In other instances the littoral forest may be deciduous, 
or dominated largely by a single species (such as Ironwood, Casuarina 
equisetifolia). The proximity of the sea is also expressed in the 
buoyancy of many of the seeds and fruits, which are commonly 
found in sea-drift, and which help some at least of the characteristic 
species to attain a very wide distribution. This is said to be the 
case with the Coconut, plants of which form such a characteristic 
feature of many tropical sea-shores (Fig. 160). 

Further Seral or Edaphic Communities 

Apart from various seral types already mentioned, such as the 
dunes and, presumably, littoral woodlands dealt with in the last 
section, and biotic plagioclimaxes which in some respects are of a 
seral nature, there are yet others to consider in tropical and sub- 
tropical regions. Outstanding are various types of forested and 
reedy swamps, secondary scrubs and forests, and weedy com- 
munities of many kinds. 

Swamps occur chiefly around the edges of quiet bodies of fresh 
water, in sheltered arms of lakes or sluggish rivers, and in filled or 
filling hollows where the ground is at least waterlogged and where 




free water accumulates on the surface for some period or periods of 
the year. Here, as in temperate regions, there is often a luxuriant 
development of largely erect monocotyledonous plants, such as 
Papyrus (Cyperus papyrus) or species of Reed (Phragmites) or Cattail 
(Typha). These form characteristic reed-swamps, with roots under 
water or in saturated soil, and with shoots extending more or less 
high into the air. Whether the inundation is permanent or periodic, 
and regardless of the water-level being relatively stable or fluctuating, 
such swamp-plants typically contain air-passages for the aeration of 
their roots and other covered parts. 

Fig. 160. — Coconut Palms along a tropical sea shore. 

In many cases, particularly in the warmer regions, the shallow 
water is colonized by shrubs or trees, which may have special aerating 
roots after the manner of mangrove types. Swampy grounds in both 
the Old and New World tropics are frequently occupied by almost 
pure stands of certain species of Palms, while even where the forest 
is mixed it is usually much less rich in species, and particularly in 
large tree species, than on drier land. These swamp-forest trees 
usually, but not always, belong to species not normally found in 
the surrounding forests. They are said in some instances, for 
example in Burma, to be bare of leaves at the height of the rainy 
season, when they stand in a metre or more of water. Like tropical 


rain forest, the swamp-forest may consist' of several tiers and may 
be plentifully supplied with lianes, that in some instances are 
described as having a short stem which reaches up only to the surface 
of the water in the rainy season, and from which arise dispropor- 
tionately long, slender shoots. 1 The numbers of terrestrial herbs 
depend upon such features as the depth and duration of flooding : 
often they are few, being chiefly members of the Sedge family 
(Cyperaceae). However, epiphytic Orchids and Ferns can be plenti- 
ful, as can Mosses and Liverworts. 

Whereas the swamp vegetation just described is evidently hydro- 
seral, exemplifying stages in the succession, from open water to 
forest, that appears to take place in the same general manner as in 
cooler regions, there is one peculiarity in the tropics, where humus 
does not normally accumulate to any great degree. This is the fact 
that the majority of tropical swamp soils are not peaty, containing 
as they do little if any more humus than soils of normal drainage. 
However, where the water is poor in dissolved mineral matter 
(oligotrophic) some peat formation can occur, leading to the develop- 
ment of ' moor-forests ', which are commonly called the tropical 
equivalent of the ' highmoors ' of cool regions. Corresponding to 
this on one hand and, on the other, to the normal (non-peaty) 
swamp soil with a relatively eutrophic water-supply, there thus 
appear, in the tropics, to be two types of hydrosere leading to 
different types of climax forest which in both cases are edaphic 
rather than climatic. For in eutrophic waters the raising of the soil 
level is due mainly to the accumulation of inorganic sediments, 
while in oligotrophic waters such raising is chiefly the result of 
accumulation of plant remains. In both cases the soil level rises 
scarcely if at all above the height of the highest water-level once 
this has been reached, as conditions for further substantial accumula- 
tion then cease to exist in the tropics where organic breakdown is 
rapid. Consequently the hydrosere appears to end with the forma- 
tion of ground in which the water-table is near the surface during 
at least part of the year. Such ground bears forest more or less 
like the climatic climax in structure, but different in floristic composi- 
tion owing to the local water conditions. It is thus an edaphic 
climax and, like the hydrosere which engenders it, exists in two 
forms according to whether the soil consists largely of deposited 
silt or peat. 

' A similar phenomenon may be observed in some areas of Amazonian rain 
forest that are subject to periodical inundation (R. E. Schultes voce). 


Particularly characteristic are the peaty moor-forests originating 
in oligotrophic waters, which are widespread in the rain-forest region 
of southeastern Asia, where they are evergreen and dominated by 
dicotyledonous trees. These last may be as much as 30 metres 
high on the edge of such areas where the climax has been reached, 
but often diminish gradually towards the centre where the vegeta- 
tion typically consists of earlier stages, including dwarf forest and 
even pools of water. Kneed and other aerating roots may help to 
make the surface of the ground an ' impenetrable ' jungle even in 
the mature forest. The dominants are often species peculiar to this 
type of vegetation ; they are relatively few in number and show a 
strong tendency to be gregarious, with a single species sometimes 
forming an almost pure community. Often associated are Palms 
and Screw-pines, with abundant epiphytes and herbaceous swamp- 
plants among the furnishings. 

In eutrophic waters the early stages clearly consist of free-floating 
' sudd ' and communities of submerged aquatics, followed, when the 
water becomes sufficiently shallow owing to silting, by rooted 
floating-leaf vegetation consisting of Water-lilies, etc. This prepares 
the habitat for emergent aquatics that soon constitute the reed- 
swamp stage, which in turn is succeeded by scrub or low forest. 
Though it may contain more trees per unit area, the (edaphic) 
climax canopy tends to be more open than in rain forest, so that 
light-loving species are commonly included in the undergrowth. 
Moreover, the total number of species per unit area of this climax 
tends to be smaller than in the rain forest, but greater than in the 
serai stages. 

It is of interest to note that not only do tropical hydroseres run 
much the same course as temperate ones, the recognizable stages 
often having a closely comparable physiognomy, but many of the 
genera involved are the same in both these main climatic zones, 
and not a few of the species are closely related or, in some instances, 
identical. This is especially the case in the early stages in eutrophic 
waters, when, as in other extreme habitats, only a few co-dominants 
or even a single dominant may prevail. Thus in Panama the 
Common Reed (Phragmites communis agg.) and/or Narrow-leafed 
Cattail (Typha angustifolia) may largely dominate the reed-swamp 
stage, as may identical or similar species over much of temperate 
North America and western Eurasia. Characteristic inhabitants of 
rocks in rushing water are representatives of the peculiar family 


Primary xeroseres in the tropics may be observed for example on 
recently emerged or volcano-devastated areas such as the East 
Indian island of Krakatau, where after three years an associes con- 
sisting of a lower layer of Blue-green Algae and an upper one of 
Ferns and other vascular plants was found to clothe the surface of 
the pumice and ash in some inland places. Eleven years later, in 
1897, the interior supported a dense growth of Grasses, with isolated 
shrubs and fairly numerous forbs as well as Ferns. There were, 
however, very few lower cryptogams to be seen apart from ter- 
restrial Algae. Nine years later still, the shore supported a belt of 
well-developed maritime woodland complete with climbers, etc., 
but the inland savanna persisted. Subsequently it developed into 
a mixed woodland of fair luxuriance, and came to show every 
indication of progressing ultimately to the local rain forest (apart 
from some floristic depauperation and, probably, slowness of succes- 
sion due to isolation and the consequent difficulties of colonization). 
Thus the usual sequence, such as we have seen elsewhere, of domin- 
ance by cryptogams and then by herbs and finally by trees, holds 
true in this tropical xerosere, though it should be noted that 
in rain-forest areas, such as this, there is a preponderance of 
phanerophytes among the flowering plants — often from quite early 

As in the case of the hydrosere, it seems that in the tropical 
xerosere the number of species goes on increasing to the end, whereas 
in temperate regions the numbers of species in both hydrosere and 
xerosere tend to rise to a maximum and then decline as the climax 
is approached, the decline usually starting when the community 
becomes closed. Other humid tropical regions appear to have 
xeroseres of a generally similar nature to that observed on Krakatau, 
even if the pioneers in some cases are forbs, Grasses, Sedges, or 
even woody plants (especially in secondary successions). 

The sea-shore and littoral communities outside the normal forest, 
already dealt with in the last section, are probably indicative of at 
least potential successions. However, with the factors of the 
environment as overwhelming as they often are in such situations, 
it seems unlikely that the successions will become actual as long 
as the shore-line continues as at present. Rather does it appear 
that each zone is in equilibrium with its particular environment. 
This, as we have seen, is apparently the case with many mangroves 
as well as with the latest stages of tropical hydroseres. But where 
accumulation can continue, as on some sandy shores, the communities 


may undergo rapid change and the zones of vegetation be actual 
stages in continuing successions. 

Very widespread in the tropics are ' secondary ' scrubs and forests 
that form parts of secondary or deflected successions engendered 
particularly by Man or his domestic animals. When derived from 
tropical rain forest, such communities are always more or less 
unstable, whether they consist of weedy herbs, scrub, savanna, forest, 
or a * chaotic wilderness of trees, shrubs, herbs and climbers '. 
When left to themselves and protected from burning, felling, and 
grazing, they are gradually invaded by primary forest species and 
proceed towards the climatic climax which with little doubt would 
ultimately be re-established. But where they are subjected to 
recurrent fires or persistent grazing, deflected successions set in and 
lead to biotic plagioclimaxes. Such are, probably, many tropical 
grasslands and savannas. Even in the forest, according to Professor 
Paul W. Richards (in litt. bints), ' too frequent cultivation, i.e. 
shifting cultivation on too short a rotation, is one of the most impor- 
tant causes of deflected successions ', and, ' especially if accompanied 
by intermittent burning and grazing, leads to the invasion of forest 
areas by savanna Grasses and eventually to the establishment of 
" derived " savanna '. 

Shifting cultivation, which is practised by native peoples in nearly 
all tropical forested areas, is by far the most important cause of 
forest destruction there. In its course the trees are felled and 
burned, after which one or more crops are raised before the fertility 
of the soil is lost by leaching, erosion, and the exposure of humus 
to the sun, whereupon the plot is abandoned and another cleared. 
The secondary succession following abandonment usually starts 
with series or quick-growing herbs and continues with more lasting 
ones. During such re-establishment of vegetation, soil fertility 
becomes partially or largely restored, so that after some years the 
' secondary ' (or tertiary, etc.) forest may be cleared and cultivated 
again. The practice is, however, liable to be extremely wasteful 
— especially when clearing and cultivation are undertaken at too 
frequent intervals. Less drastic in their effects are the abandonment 
of plantations and selective exploitation of timber, while strong winds 
may also fell trees and start secondary successions. On the other 
hand, shifting cultivation that is not too intense and short in rotation 
is not necessarily wasteful, and may be preferable to some forms of 
what is intended to be permanent cultivation, because it is less 
destructive of soil fertility. It may also be less conducive to erosion. 


Secondary forest tends to be lower and to consist of trees of 
smaller dimensions than does primary forest, the young growth being 
often remarkably regular in including even-aged stands of one or a 
few woody species, whereas later on the growth may be extremely 
haphazard as already indicated. The early dominants are commonly 
light-demanding and ' weedy ', as are the associated herbs. Such 
forest may also be recognized by its floristic composition, which 
usually differs markedly from that of the primary forest, even though 
there are probably few if any species entirely restricted to the former. 
Many of the components of the secondary forest are unusually wide- 
spread, some often being introduced aliens, while its trees are mostly 
quick-growing (e.g. 12 metres in three years), short-lived, and pos- 
sessed of efficient means of seed-dispersal. Their leaves tend to be 
of more various sizes and shapes than those of rain-forest trees. 
The most shade-intolerant and quickly growing species are, as might 
be expected, most characteristic of the early stages of the secondary 

The secondary forest at least when young is often dominated by 
a single or small number of species, and usually has a much smaller 
flora than the primary forest ; when very old, however, it may be 
indistinguishable from virgin forest. On the other hand with long- 
continued grazing, mowing, or recurrent burning, secondary savanna 
or grassland is commonly formed, often characterized by species 
of Lalang Grass (Imperata) ; alternatively, as in temperate grass- 
lands, there may be still further regression with overgrazing, or 
invasion by Bracken (Pteridium). In general, however, secondary 
successions appear to reproduce in their later stages the changes 
characterizing natural regeneration of primar\ forest, in which gaps 
formed by the death of large old trees are first filled by the easily 
dispersed and fast-growing dominants of the early stages of secondarv 
forest. The earlier stages in larger clearings commonly include 
quick-growing Grasses and other weeds which are characteristic of 
disturbed tropical areas, though they may be even less lasting than 
their counterparts in cooler regions. 

While shrubs may form a stage in the successions occurring in 
rain-forest areas, often they are omitted, dominance passing directly 
from herbaceous plants to trees. However, in drier regions shrubs 
are apt to be important in the secondary successions, sometimes 
remaining as more or less lasting dominants in what appear to have 
been previously forested (though ' marginal ') areas. Indeed vegeta- 
tional differences, such as frequently arise from differences in the 


soil, tend to be much more marked in the dry districts of the tropics 
than where rain forest prevails. Outstanding are the laterite soils, 
typically reddish in colour owing to ferric compounds, which are 
extremely poor in alkalis and nutritive salts as well as in water- 
retaining capacity. They are consequently unfavourable to most 
plants and support relatively poor vegetation — examples being the 
forests in Burma dominated by Eng (Ira), Dipterocarpus tuberculatus, 
which often forms almost pure consociations and alone grows up well, 
the other trees being stunted and gnarled. Similar poor forests 
may also be found on light sands, ' bare ' limestone, and dry ridges 
of acid-weathering rocks, though these areas often support no more 
than thorn- woodland or even scrub. Such vegetational poverty is 
usually in part engendered by the relatively little humus-accumula- 
tion, due to rapid breakdown in the tropics. In other cases porous 
siliceous soils may be occupied by forests having a particular char- 
acter owing to dominance by particular plants, such as Sal-tree 
(Shorea robusta) or various Bamboos. Communities of these last 
are often virtually pure, containing no associates apart from small 
cryptogams, and in many cases apparently owe their origin to 

Altitudinal Effects 

The vegetation-types of high altitudes above the tree-limit in 
tropical as well as other regions were covered in a general way in 
the last chapter, and the communities of fresh and salt waters are 
treated in Chapters XV and XVI, respectively. But here we must 
consider briefly the upland types occurring below the timber-line 
in the tropics and subtropics. 

The basal zone of a range of mountains has in general a greater 
rainfall than the neighbouring lowlands, being consequently often 
occupied by communities resembling the relatively moisture-loving 
ones of the lowlands. This is true in tropical regions where, 
accordingly, rain forest is very widespread and frequently very 
luxuriant on the lower slopes of mountains. Above comes the 
montane zone, where the precipitation is often phenomenally high, 
and which in the equatorial region is still tropical in its lower levels ; 
but at higher levels here, and throughout its altitudinal range to 
the north and south, the montane zone is rather temperate in type, 
with vegetation corresponding more to temperate rain forest. Thus 
the dominants are often of particularly massive growth and rich 
branching, but devoid of plank-buttresses ; they are evergreen but 

4 6 8 



tend to have smaller leaves than those of" the tropical rain forest. 
The general foliage, too, is less dense, and commonly only two tree 
strata are discernible, allowing light to penetrate and plentiful 
ground-vegetation to develop (Fig. 161). A fair amount of humus 
may accumulate but stranglers are absent. 

Fig. 161. — Two-storied montane rain forest at an altitude of 740 metres in the 

Philippine Islands. 

Although more and more temperate species enter as we ascend 
in the upper forested zones on tropical mountains, the total flora 
tends to decrease. But whereas this decrease is particularly marked 
in the case of trees, it is not accompanied, as it is in unfavourable 
lowland situations, by any marked tendency to dominance by single 
species. Leaves in general become fewer in number and usuallv 
more slender than at the lower levels, and the epiphytes are usually 
smaller, almost all herbaceous, and mostly limited to Ferns and 
Bryophytes or more lowly cryptogams. Small climbers as well as 
epiphytes are, however, often abundant in the two-storied upper 
montane forest, as shown in Fig. 162. Epiphytic ' mosses ' 
(mostlv leafy Liverworts) tend to be particularly numerous and 
luxuriant where mists prevail in these and any still higher forests, 


characteristically blanketing the trunks and also hanging from 
almost every possible point, so that they create far more of a ' show ' 
than in the three-storied rain forest — hence the name of ' mossy 
forest ' frequently applied to such types. 

Fig. 162. — Epiphytes on trunk of tree near upper limit of montane rain forest 
in the Philippine Islands. 

Often the subalpine zone is little marked, except by reduction 
in the size of the trees and in their foliage which may gradually 
acquire a morexeromorphic structure, or, in some cases, be deciduous. 
In time, as we ascend, only a single storey is left, corresponding to 
the lowest tree one of the tropical rain forest. Often it is exceedingly 
mossy. As the trunks of the trees become shorter and relatively 
thicker, the branches tend to enlarge, at least in proportion, until 
growth becomes irregular as elfin wood is reached. In this the 
trees are twisted and stunted, being especially low and grotesque 
in the extreme forms known as ' Krummholz ', though still commonly 
festooned with Mosses etc. (Fig. 163). This elfin wood and finally 
Krummholz marks the termination of the forest and the beginning 
of the treeless alpine zone which is vegetated by scrub, tundra, and 


heathy or herbaceous vegetation, with ultimately, far above, sparse 
fell-fields, etc., as described in Chapter XIII. Sometimes, as in 
the tall mountains of New Guinea, other zones are interpolated. 

In tropical and subtropical regions of dry climate, as in temperate 
lands, forest (or intermediate savanna, etc.) may appear only in the 
montane zone — or occasionally not at all, as on the western slopes 
of the Andes in parts of South America, where scrub, steppes, or 
arid ' punas ' characterized by large cushion-plants prevail. Near 

Fig. 163. — Mossy elfin forest near summit of mountain, Philippine Islands. 
Trunks, branches, and aerial roots of trees are covered with festoons of Mosses. 

the latitudinal limits of the subtropics, something approaching 
deciduous summer forest and, above it, forest characterized by 
evergreen Conifers, may in some places constitute the upper limits 
of arborescent vegetation and simulate the march into higher 
latitudes (cf. Fig. 83, B). Details also vary elsewhere in many other 
ways, such as the altitudinal limits involved, though these last tend 
to become depressed with increasing latitude, proximity to coasts, 
or exposure to prevailing winds. However, owing to the ' height ' 
of the sun and the wide angle of incidence of its downpouring rays, 
the influence of aspect tends to be far less marked in tropical than 
in temperate and polar regions. 

14] vegetational types of tropical lands 471 

Further Consideration 

Although it is in the tropics that the most luxuriant and complicated 
of all vegetation occurs, most of the pertinent literature is again tiresomely 
scattered. For such subjects as it covers, however, this is happily 
remedied by the first book cited below ; the others are useful for further 
details or vivid illustrations : 

P. W. Richards. The Tropical Rain Forest : an Ecological Study 
(Cambridge University Press, Cambridge, Eng., pp. xviii -f- 450, 
1952). Brings together the available knowledge concerning tropical 
rain forests and related topics. See also the more general works of 
Schimper, Faber, Hardy, Campbell, Tansley & Chipp, and Newbigin 
cited at the end of Chapter XII. 

W. A. Cannon. Botanical Features of the Algerian Sahara (Carnegie 
Institution, Washington, D.C., Publication No. 178, pp. vi + 81 and 
36 plates, 1913). 

W. A. Cannon. General and Physiological Features of the Vegetation 
of the more Arid Portions of Southern Africa, with Notes on the 
Climatic Environment (Carnegie Institution, Washington, D.C., 
Publication No. 354, pp. viii +159 and 31 plates, 1924). 

J. S. Beard. The Natural Vegetation of Trinidad (Oxford Forestry 
Memoirs, No. 20, Clarendon Press, Oxford, map and pp. vi -f- 7-152, 

G. M. Roseveare. The Grasslands of Latin America (Imperial Bureau 
of Pastures and Field Crops, Aberystwyth, Bulletin No. 36, pp. 
1-291, 1948). 

R. E. Holttum. Plant Life in Malaya (Longmans, London etc., pp. 
viii + 254, 1954). 

Concerning special tropical items or regions there are, in addition to 
the works already cited, numerous and usually well-illustrated papers in 
the Journal of Ecology, which has been published continuously since 1913, 
and, almost always in German, in the 26 volumes of Vegetationsbilder 
published between 1904 and 1944 and in the long series of monographs 
edited by A. Engler & O. Drude and entitled Die Vegetation der Erde 
(Engelmann, Leipzig, 1896 onwards). 

Chapter XV 


Reed-swamp and other semi-aquatic types of vegetation in which 
at least half of the plant body is aerial have already been treated 
in Chapters XII-XIV, dealing with the terrestrial vegetation of dif- 
ferent climatic belts. This leaves the more fully aquatic freshwater 
communities, with some others, to be dealt with in the present 
chapter, followed by the marine ones in Chapter XVI. Such a 
separation of aquatic from terrestrial habitats and attendant vegetation 
seems the more proper when we reflect that whereas on land it is 
the climatic factors which are of primary importance in determining 
the distribution of particular vegetation-types, in aquatic media it 
is rather the chemical composition that is fundamental in this respect. 
This is particularly the case with salinity, which gives us our primarv 
division into fresh and salt waters. Yet the ecological differences 
between terrestrial and aquatic habitats are largely matters of degree, 
chemical and physical conditions in the soil being still extremelv 
important in the former, for example, and light and temperature in 
the latter. Even the distinction between fresh and salt waters is 
incomplete, as there are various intermediate ' brackish ' waters of 
varying degrees of salinity and, in addition, inland salt-lakes ; these, 
however, except in such instances as the Caspian Sea which are of 
marine origin, seem best dealt with in the present chapter, leaving 
only marine types to be considered in the next. 

Some Features of the Freshwater Aquatic Environment 

The temperature and some other conditions vary less in aquatic 
than in terrestrial habitats, water acting as an effective ' damper ' 
on changes concerning heat. Moreover, major bodies of water 
exercise an equalizing influence on the temperature of adjacent air. 
As a result of the minimization of variation in aquatic environments 
and of the fact that, obviously, they offer no problem of water-supply, 
aquatic plants and vegetation-types tend to be more widespread than 



terrestrial ones. However, a marked vertical distribution occurs, the 
vegetation in sufficiently deep waters being divided into zones ac- 
cording to the different depths. In this delimitation light is usually 
the main factor, though the temperature of water-masses and the 
local chemical composition and especially aeration of the water may 
also be important. The zones, at various depths in water, that are 
characterized by different forms or abundance of life, largely repre- 
sent stages of decreasing intensity of light. They range from a 
relatively bright surface ' euphotic ' zone in which the light is suf- 
ficient for the normal development of large plants, through a dim 
' dysphotic ' zone in which photosynthesizing small Algae and even 
Mosses may still flourish, to a dark and relatively deep ' aphotic ' 
zone in which only non-photosynthesizing organisms can exist. 
Owing to the varying turbidity of waters due to suspended particles, 
and to the different penetration of the sun's rays at different angles, 
the limits of these zones, which are themselves imprecise, lie at very 
different depths in different instances. 

In summer, lakes in temperate regions experience a marked rise 
in temperature, particularly on and near the surface, but this is 
followed by a decrease in autumn. Such fluctuations engender 
vertical convection currents and eddies which lead ultimately to a 
temperature of low value throughout, while at the same time they 
are important in aerating the deeper layers {see p. 477, and cf. Fig. 
165). These last may vary very little (less than 4 C.) in temperature 
between summer maximum and winter minimum. Tropical lakes 
may also show a surface temperature fluctuation with changing 
weather, but here a reduction in temperature of merely 1-2 C. is 
reported to bring about a circulation similar to that which is effected 
in cooler regions by a winter cooling of some 20 C. 

Limnology is the study of inland waters, including the environ- 
ment and all inhabiting organisms and their interrelationships. In 
it a distinction is made between the benthos (of organisms growing 
on or in the bottom material ; these may be described as 'benthic'), 
the freely floating plankton, and the swimming nekton. The dis- 
tinction is, however, somewhat artificial as many organisms are 
border-line cases. The plankton and benthos of inland waters are 
said to be limnetic, in contradistinction to those of the open sea 
which are said to be pelagic. These terms are not much used in 
limnology, where the region of open water is commonly distin- 
guished as the pelagial. In shallow coastal waters the plankton 
is apt to be mixed with forms belonging to the benthos, and 


such waters and their plankton may usefully be described as 

Halophytes are plants which can tolerate a considerable degree of 
salinity. But whereas the land halophytes and those of brackish 
waters are usually euryhaline (that is, able to tolerate a wide range of 
salt-content in different soils), most aquatic halophytes are more 
stenohaline (capable of tolerating only a narrow range of salt-content 
in different circumstances, their minimum, optimum, and maximum 
being relatively close together). Similarly, euryphotic and eurythermic 
plants are those tolerating a wide range of light and temperature, 
respectively, and stenophotic and stenothermic plants are those which 
tolerate only a narrow amplitude in such respects. 

In general the chief boundaries between different types of aquatic 
vegetation are determined by factors comparable with those operating 
on land, though the emphasis is often changed. Thus temperature, 
salinity, and light are of obvious importance in various ways, as are 
movements due to surf and currents. Light-penetration depends 
greatly on various factors such as cloudiness of the water (due to 
suspended bodies whether living or dead), reflection from the surface, 
latitude and the consequent angle of incidence of the sun's rays, 
and content of dissolved substances. Moving water, besides being 
better aerated, demands of plants mechanical qualities differing 
from those required in still water, and, moreover, stagnant fresh 
water tends to have vegetation of very different composition from 
running water. Quite apart from this, shelter from currents and 
waves can be an important factor and even a major necessity in 
aquatic media. Currents, on the other hand, can disperse plants 
and improve such conditions as aeration and the potentialities for 
nutrition. Rainfall tends to be of significance chiefly in affecting 
the salinity of lagoons, particularly in wet tropical regions. Aquatic 
plants inhabiting such waters and the mouths of many streams must 
be widely euryhaline ; thus certain Diatoms living in waters where 
the salt-content changes widely and rapidly are able both to absorb 
and let out salt equally quickly, according to the concentration of 
the medium. 

We have seen that light-penetration in water is a very variable 
factor. This is important because different Algae and other aquatic 
plants needing light for photosynthesis vary greatly in their actual 
light-requirements. As explained in the next chapter, this variation 
is in part associated with the predominating colours of the different 
groups, but it also exists in fresh waters where Red Algae are usually 


little in evidence and Brown Algae are practically unknown. Thus 
different species are differently adjusted as to both wave-length and 
intensity of light, the green surface-forms flourishing where the 
intensity is high and red rays plentiful, whereas they are hardly able 
to utilize the green and blue rays which penetrate deeply into clear 
waters. (The still shorter-length ultra-violet radiations, which may 
be lethal, penetrate very little.) Photosynthetic activity of higher 
plants also diminishes downwards, though the depth at which the 
daily assimilation barely compensates for respiration varies greatly 
with different species. This ' compensation-point ' for the Canadian 
Water-weed (Elodea canadensis) is about 10 metres, whereas for the 
Moss Fontinalis it is about 18 metres under comparable conditions. 

Highly important is the physical nature of the substratum, benthic 
vegetation assuming a very different character according to whether 
the ' bottom ' is rocky, gravelly, sandy, or muddy — in particular, 
whether it is hard or soft. Whereas rocky beds are often suitable 
for attachment of Algae, soft substrata are favoured by most higher 
plants that have to take root. In the intimacy of smaller bodies of 
fresh water, the chemical nature of the substratum tends to assume 
a greater importance than in large and deep lakes. This is because 
the substances in solution exert an influence largely according to 
their concentration. Thus, the flora and vegetation often differ 
markedly according to whether the water is rich or poor in dissolved 
calcium carbonate and some other salts, while the relative abundance 
of organic materials or humus may also be important. Ice action 
is significant on many polar and other frigid lake-shores as well as 
sea-shores, and even in temperate regions can profoundly affect the 
nature of the surface and the composition of the marginal flora. 
Moreover, extensive freezing of water may significantly alter the 
acidity and nutrient and other chemical content of the underlying 

As regards periodic phenomena, temperature in general exhibits 
far smaller and less rapid fluctuations in water than in the air, and 
is therefore less influential than on land. But whereas perennial 
marine Algae even in cold regions may exhibit no period of winter 
rest, seasonal differences in both quality and quantity of vegetation 
tend to be well marked in small bodies of inland water. This 
commonly results from the considerable variation in temperature and 
easy formation of ice, while especially in boreal and austral regions 
seasonal variations in light may cause a distinct periodicity. More- 
over, the amounts of nitrate and phosphate available in bodies of 


water may also fluctuate seasonally to a marked extent, which again 
may be the basis of alterations in the inhabiting population. It is 
often contended to be as a result of the prevailingly low temperatures 
that cold-loving Diatoms predominate in the lakes of central Europe 
in winter and spring, followed by Peridinians in summer and by 
Cyanophyceae when the temperature has reached 20 c C. However, 
as pointed out by Mr. Robert Ross (in litt.), many other Diatoms are 
by no means cold-loving, and it may well be that the control of this 
cycle is more chemical than physical, silica depletion for example 
being very important. Hot springs, having a constantly high tempera- 
ture, support a largely peculiar and often very limited flora 1 — as do 
snow and ice at the lower end of the scale of temperatures allowing 
plant activity (see pp. 489-91). 

Although the food requirements of Algae and many other water- 
plants are still but poorly understood, it is clear that the presence 
or absence of certain mineral salts, derived from the substratum or 
inflowing currents, is of outstanding importance in helping to deter- 
mine both the composition and the luxuriance of the vegetation 
developed. Quite apart from questions of salinity, small freshwater 
lakes are particularly dependent on the chemical and physical nature 
of their beds. As already explained in Chapter XI, chemical 
' reaction ' (acidity, neutrality, or basicity) and especially nutritive 
salt-content are of fundamental significance in determining details 
of development of aquatic vegetation — especially plankton — and, of 
course, also to the dependent animal populations, in bodies of fresh 
water. Hence the classification of the latter largely on the basis of 
productivity into ' oligotrophic ' (poor in nutrients, with a hard rocky 
bottom and rapidly deepening water), ' dystrophic ' (also poor in 
nutrients but rich in humus and acidic in reaction), and ' eutrophic ' 
(poor in humus though commonly silted and shallow, rich in nutrients 
including combined nitrogen, phosphorus, and often calcium). Par- 
ticularly are the nitrate and phosphate contents of decisive importance 
in the matter of biological productivity in fresh waters as well as 
salt ones. Also of great significance are the contents of dissolved 
oxygen and carbon dioxide, which vary at different depths and in 
different seasons — cf. Fig. 165 for oxygen fluctuations. 

In oligotrophic lakes, cold-water Fishes such as Trout are often 
plentiful ; these lakes commonly show succession towards the eutro- 
phic type. Dystrophic lakes, on the other hand, usually lack these 

1 For example in hot springs above 45° C. only Schizophytes appear able to 
persist, as indicated on p. 499. 


deep-dwelling cold-water Fishes and sometimes other types too, their 
fish-productivity being poor at best, while succession proceeds to 
peat bog. Eutrophic lakes usually also lack deep-dwelling cold- 
water Fishes, though they are often suitable for Perch, Pike, Bass, 
and other warm-water Fishes ; succession in them is to swamp or 
marsh. Oligotrophic and dystrophic waters are often rich in 
Desmids, eutrophic ones in Diatoms and Blue-green Algae. 

Among the various ways in which living organisms alter fresh 
waters is in the matter of gas-content. In general, green plants 
(except in non-photosynthetic periods) remove carbon dioxide and 
add oxvgen, while animals do the reverse, so that in the upper, well- 
lit lavers there tends to be a superabundance of oxygen and in the 
deeper and darker layers more carbon dioxide than above. Such 
considerations lead to the recognition of two types of waters, namely, 
those where the gas-content is almost constant in all layers, and those 
where it decreases markedly in the depths. The former are chiefly 
masses of water in which vertical currents cause almost constant 
mixing. In the latter it is common to recognize in summer in tem- 
perate regions (1) a wind-stirred and largely homogeneous ' epilim- 
nion ' or surface layer rich in oxygen (because of contact with the 
air as well as photosynthesis), usually extending to a depth of 10-15 
metres ; (2) a middle ' metalimnion ' or ' thermocline ' where the 
temperature and oxygen-content decrease rapidly ; and (3) a * hypo- 
limnion ', underneath, where the water is virtually stationary and no 
oxygen enters from above. Tropical lakes commonly differ from 
those of temperate regions in that the shallower ones ' only stratify 
for short periods, if at all, while deeper ones may have cyclical 
stratification or, if very deep, e.g. Nyassa and Tanganyika, may turn 
over very rarely or not at all ' (R. Ross in litt.). 

It is particularly in lakes sheltered from wind-disturbance in 
temperate regions that the oxygen stratification tends to follow the 
bottom contours, and the surface-waters in periods of quiescence are 
often more alkaline than deeper ones. However, in autumn the 
surface-waters cool and sink, carrying down dissolved oxygen, and 
the deeper masses rise to take their place, so getting aerated by a kind 
of reshuffling each year. The greater the amount of nutrient material 
and accordingly of organic life in a lake, the faster does the oxygen 
disappear in the depths during the quiescent period of summer. 
Carbon dioxide, being complementary in metabolism, exhibits a 
largely reverse trend, disappearing from the well-lit upper layers but 
accumulating below. In spite of the relatively small amount in the 


free atmosphere, which averages about 0.03 per cent, of carbon dioxide, 
this gas, being easily soluble in water, is widely abundant in lakes. 
Thus many contain more than 20 c.c. per litre in the depths, although 
in oligotrophic lakes the content may be as low as 1 c.c. per litre. 
Oligotrophic lakes may also show little variation in the oxygen content 
at different depths, in marked contrast to eutrophic ones. 

The so-called ' lime ' content (particularly of calcium carbonate 
and bicarbonate) of freshwater lakes and streams varies greatly, 
peaty waters being especially lime-poor and ' soft ', whereas those 
originating in calcareous districts are mostly lime-rich and ' hard '. 
Variations also occur at different depths and times of the year : 
thus in summer periods of relative stagnation, living organisms 
may remove much of the ' lime ' from the upper layers, while the 
deeper ones, which are already rich in carbon dioxide, actually be- 
come enriched in lime. Calcium and allied carbonates and bicarbon- 
ates have a marked ' buffering ' effect against changes in the reaction 
or pH level of a body of water, and as the reaction (or some condition 
associated with it) is often important in affecting the flora and conse- 
quently the vegetation, so are such ' salts ' important. Most favour- 
able is a weakly basic to neutral reaction, markedly acidic waters 
being biologically unfavourable : hence the limited and peculiar flora 
of scarcely buffered bog-waters, the acidity of which is associated 
with marked poverty in lime. 

With regard to the impress of the environment as a whole, there 
is insufficient data as yet to compare the vegetational productivity 
of different climatic zones. Thus although some tropical inland 
waters may be more prolific as producers of plant or animal life than 
some extra-tropical ones, others are practically barren. Many extra- 
tropical lakes naturally rank extremely high in productivity, and for 
many which do not so rank a great deal can be done by the addition 
of fertilizers or by such manuring as is, for example, practised in 
European carp-ponds. For productivity of lakes is related largely 
to such factors as chemical content, turbidity, and light. Even in 
the Arctic the present writer has collected samples from small lakes 
and ponds that have shown a surprising wealth of algal forms : in 
one instance of six samples taken in as many small vials from shallow 
pools and peaty puddles in Baffin Island in July and August, 1936, 
no less than 179 different species or varieties of Algae were determined. 
Nearly all of these were microscopic, a large proportion being Des- 
mids. Moreover a considerable number of organisms, such as the 
familiar planktonic Dinoflagellate Ceratium hirundinella and species 


of the Entomostracan genera Daphnia and Cyclops, range from the 
polar regions to the tropics, so suggesting again that at least those 
conditions which are critical for them are relatively uniform over 
remarkably wide areas. 


Planktonic organisms are those which float freely on or in a body 
of water ; they may be roughly divided into animal and plant types, 
constituting, respectively, zooplankton and phytoplankton. The main 
categories of freshwater plankton are (£) the ' limnoplankton ' of 
lakes and ponds ; (ii) the ' potamoplankton ' of slow streams and 
rivers ; and (tit) the ' cryoplankton ' of lasting snow, neve, and ice. 
The cryoplankton is so distinct in habitat and form that it seems best 
treated separately in the next section. In addition we may distin- 
guish in lakes and ponds (iv) the ' tychoplankton ' of forms transported 
from the littoral or from affluents by currents. 

Planktonic organisms must be able to remain suspended in water, 
and this they do either by having the power of active locomotion, 
particularly by flagella, or, more often, in the case of phytoplankton, 
by having suitable ' form-resistance '. This latter is commonly 
expressed in projections from the surface of the usually minute body 
and, in addition, for success requires a specific gravity near that of 
the surrounding medium. Thin, light, and flat or curved cell-walls, 
such as are often found in planktonic Diatoms, help considerably in 
this connection ; so do gelatinous sheaths possessing almost the 
same density as water, and light food-reserves of oil, or, of course, 
still lighter bubbles of gas. Very important is smallness of size, 
for we all know that large bodies sink more rapidly than small ones 
of the same material. Also significant is the specific surface area, 
which is the ratio of the total surface area to the volume of the body ; 
for the larger the ratio, the greater will be the friction caused during 
sinking, and consequently the slower this last will be. Hence the 
frequent provision of spines, horns, ridges, and the like on the out- 
side of planktonic organisms, such as are illustrated in Figs. 5 (note 
especially the forms of Desmids), 6, and 7. Moreover, owing to the 
greater viscosity of water as well as protoplasm at low temperatures, 
the ability of planktonic organisms to float tends to be greater during 
winter than summer. All such ' flotation-adaptations ' are, however, 
unable to prevent fairly rapid sinking of most except the smallest 
among the non-motile planktonic organisms if the water is entirely 


static. Rather is it the turbulence of the water caused by the eddy- 
diffusion currents that maintains a state of continuous mixing par- 
ticularly in the epilimnion, and keeps in this upper layer a sufficiency 
at least for survival of the non-motile phytoplankton which chiefly 
flourishes here. 

Whereas in the sea the differentiation of open-ocean (pelagic) and 
near-shore (neritic) planktonic organisms is often marked, so that 
the two communities may be very different, in inland lakes the dis- 
tinction is relatively poor. Indeed, close relatives of nearly all the 
species of phytoplankton here occur also in the littoral, whence the 
open waters of the pelagial were evidently colonized. Moreover, 
the number of truly planktonic species in lakes is relatively small, a 
large proportion of the types found being really ' tychoplankton ' 
(see above), which have scarcely more claim to membership of the 
community than have the particles of inorganic and dead organic 
suspended matter (' tripton ') present. Still, the numbers of actual 
' plankters ' (planktonic individuals) may be great, especially when 
they exist in highly profuse ' blooms '. 

Freshwater plankton communities are composed of (a) producers 
and (b) consumers of organic matter, the producers being almost 
entirely chlorophyll-containing autotrophic plants. The consumers 
or non-producers, including normal Bacteria and Fungi which make 
up most of the so-called ' saproplankton ', are dependent upon the 
carbohydrates and fats and proteins synthesized by the chlorophyll- 
bearing plankters. In general the phytoplankton is composed not 
only of the relatively large and obvious types which dominate the 
so-called ' net-plankton ' and include most of the Diatoms and Dino- 
flagellates, but also of extremely minute ' nannoplanktonic ' (micro- 
planktonic) species which pass through even very fine nets (of No. 25 
silk bolting-cloth) and are usually obtained by centrifuging. In 
lakes, from one to a very few species commonly predominate and 
make up the vast bulk of the phytoplankton at any one time, and 
these may include minute nannoplanktonic types particularly of 
Peridinians or other flagellates (e.g. Rhodomonas lacustris). Bacteria 
normally contribute only a small proportion of the total ' biomass ', 
and so may even the large species that dominate the net-phyto- 
plankton. In the zooplankton, on the other hand, such large types 
as Daphnia tend to be most prominent. 

In general the most plentiful organisms in freshwater phyto- 
plankton are unicells or small colonies — whether flagellated or 
non-motile — the Bacteria, Cyanophyceae, Chlorophyceae (including 


Desmidiales (Desmids) and unicellular as well as colonial Volvocales), 
Bacillariophyceae (Diatoms), Dinophyceae (Dinoflagellates), and 
several other groups of flagellates, etc., being commonly represented. 
During spring or summer maxima a greenish, yellowish, or brownish 
1 soupiness ' may be evident to the naked eye, the composition and 
luxuriance of the community being very variable in time and space. 
Thus from an aircraft the colours and general appearance of the 
water of even adjacent tarns may be strikingly different, especially 
in boreal regions. Except in the tropics where filamentous forms 
are sometimes dominant (R. Ross in litt.), macroscopic plants are 
commonly lacking in the real plankton, as are Rhodophyceae and 
Phaeophyceae, though some Fungi may occur. 

Continuous investigation of a lake or pond over a period of years 
is likely to reveal striking changes recurring seasonally in much the 
same combination and form each year. This variability in type and 
abundance at different seasons, or periodicity, as it is called, is shown 
by (1) perennial species which occur in different densities at different 
times, and (2) ephemeral types which spend the rest of the year 
in some resistant stage usually on the shore or lake-bottom. The 
fluctuations are due to interaction between the rates of multiplication 
and of depletion, the former being dependent on basic biotic in- 
fluences and the latter largely on natural mortality, predators, and 
mechanical factors such as sedimentation. Nor do spring forms 
commonly recur in autumn, owing to the different light and tem- 
perature relationships ; for in other than tropical regions these two 
leading factors exhibit marked and important differences in lakes at 
different seasons. 

The circulation of water in winter and early spring brings up 
nutrients to the surface layers. This often leads to an early ' bloom- 
ing ' of Diatoms. But with the onset of summer stratification, 
accompanied by a profuse growth of Green and other Algae, the 
stock of nutrients cannot be supplemented sufficiently to maintain 
abundant new growth and reproduction. This is because most of 
the nitrate and phosphate ions, particularly, have already been re- 
moved from the upper layers and stored in the bodies of the organisms 
that flourish there. Certain Peridinians which can manage with a 
minimum of inorganic nutrients are then apt to appear in fair num- 
bers. However, in late summer vertical convection gradually ex- 
tends deeper, causing a replenishment of the upper layers from the 
nutrient-rich water of the depths, so that population expansion can 
again take place — cf. Fig. 165. The cycle is perennial and more or 


less perpetual in that the depths are all the time enriched by a ' rain ' 
of dead bodies containing the all-important nutrients. 

It is chiefly in late summer and autumn, when there has been an 
extensive depletion of mineral nutrients but some replenishment 
and, meanwhile, a copious increase in organic substances, that 
water-blooms of Cyanophyceae occur. Then again in autumn there 
may be another Diatom ' maximum '. To the extent that each new 
population appears only after the requisite conditions have been 
provided by its predecessor, there is here a kind of successional 
sequence, although actually such phytoplanktonic stages are probably 
all proseral in being non-essential to, or at all events not forming 
part of, the autogenic main sere. 

In connection with the spatial distribution of plant communities 
which is the mainstay of our subject, we should recall that each 
physiological activity, such as photosynthesis and reproduction, is 
greatly affected by various conditions of the environment. Usually 
with each such ' function ' there is for every pertinent environmental 
factor a minimum below which, and a maximum above which, there 
is no activity ; somewhere between lies an optimum at which the 
function involved is carried on best. These ' cardinal points ', how- 
ever, may vary with other environmental conditions, even as they do 
of course with different organisms. With such rare exceptions as 
perspiration, which generally increases with increasing temperature 
until death from overheating occurs, each physiological function 
responds in this manner. So does the organism respond as a whole 
to change in an external factor — hence the importance of physio- 
logical considerations in plant geography. But because several fac- 
tors normally change at one time, and indeed go on changing all the 
time, the effects are exceedingly complex and usually difficult to 
analyse. Moreover the demands and reactions of various types, 
species, and even lower entities or different stages of plants are 
themselves extremely various. 

In water, as we have already seen, some of the environmental 
factors which are most variable on land are damped down, but 
others retain a strong hold, as it were, on plant activity and distri- 
bution. Furthermore, the planktonic population depends not merely 
(and obviously) on the systematic groups present and able to grow 
and maintain life in the face of often unfavourable conditions, but 
also on the rate of reproduction and depletion of the component 
forms. Such depletion may be more rapid than in most other 
types of plant communities, and numbers may fluctuate greatly 


because of sinking, death, removal by currents, and consumption by 

In lakes there is commonly a marked and steep vertical gradient 
of phytoplanktonic distribution, especially when the body of water 
is limited in extent. For example, at the surface we may find an 
almost continuous investment of often quite large plants such as 
Duckweeds (Lemna spp.), Water-hyacinth (Eichhornia crassipes), and 
the types shown in Fig. 164. Whereas these macroscopic plants 
might be considered as belonging to the littoral, they often occur as 
pelagials on ponds and small lakes, especially in warm regions. 
Such relatively large floating material comprises the pleuston (hemi- 
plankton), which is commonly defined as consisting of macroscopic 
plants and as including those floating freely within the body of water 
as well as on its surface. Furthermore, there are some microscopic 
floating organisms (neuston) which stabilize their position upon the 
surface of quiet water by employing surface-tension. For example, 
a Green Alga, Nautococcus sp., becomes attached to the upper 
surface-film by means of a flotation disk — thus essentially living as 
an aerial organism, and forming conspicuous water-blooms of dry 
(powdery) appearance. Other neuston organisms hang down in the 
water, from the surface-film. 

Within the body of the water, most of the phytoplankton is usually 
concentrated in the top 10 to 15 metres, its permanent survival being 
limited to depths where more food material is made by photosynthesis 
than is used in respiration over an average 24-hour period. This 
maximum depth is dependent on light-penetration ; even in the 
clearest alpine lakes the layer of water that is at all densely populated 
by phytoplankton scarcely exceeds 50 metres in thickness. Indeed 
in the majority of lakes, at least in the higher latitudes, most of the 
phytoplanktonic life is concentrated in the uppermost 5 metres 
(according to Professor G. W. Prescott in litt.), while at depths below 
about 30 metres the numbers of individuals decline to very small 
values. Yet within these limits different species vary greatly in their 
preference. Thus more than half of the forms are usually concen- 
trated in the uppermost 10 metres or less, while others attain their 
maximum concentration at or below a depth of 10 metres. Not a 
few types, such as Cyanophyceae provided with gas-vacuoles, have 
a specific gravity of less than unity and so are concentrated at the 
surface. On the other hand some phytoplanktonic species can have 
their maximum density at depths greater than 30 metres, an example 
being the Diatom Asterionella formosa in some circumstances in 




Fig. 164.— Vascular plants floating freely on fresh water. A, a Bladderwort 
(Utricular id), showing underwater branches and leaves bearing bladders which 
trap minute aquatic organisms. B, another free-floating aquatic plant, Pistia 
stratiotes, that has roots and is very widely distributed in the tropics and sub- 
tropics. (X £.) C, a ' Batrachian ' Ranunculus, R. aquatilis s.l., floating in the 
surface water of a pond near Babylon, Iraq. The finely dissected leaves are 
immersed but the flowers rise slightly above the surface, being photographed 

from above. 

Europe. Such tendencies result in largely different communities at 
different depths — even in the same column of water at a particular 
time. This is illustrated by the diagrams A (representing Algae 
other than Diatoms) and D (representing Diatoms) in Fig. 165, the 
numbers being those of organisms per litre in a Wisconsin lake, and 
the 6 component parts of the figure being taken at approximately 
monthly intervals during May to October. 

The conditions bringing about these varied types of depth-distri- 
bution in phytoplankton appear to be those whose cardinal points 
limit physiological activity, the depth of greatest population-density 
being that of optimum conditions (the resultant of the factors involved). 
As we have seen, these optimum conditions vary greatly for different 
organisms. The issue is, however, rendered uncertain by the 
operation of mechanical factors such as the dynamics of water. 
Indeed the most important agent affecting the distribution of plankton 




in general is the movement of the water— particularly the mixing 
action of eddy-diffusion currents. Non-motile forms are kept sus- 
pended primarily by these, and under conditions of active eddy- 
diffusion are distributed more or less throughout the zone of turbu- 
lence — apart from a tendency to decrease just below the surface, 

MAY 22 

N C A 

50 10 1100 

JUNE 18 

JULY 20 
R N C A 

17 10 8 3400 

AUG. 24 
R N C A 

27 4 50 4100 


SEPT. 14 

R N C A 

340 5 25 4300 


Fig. 165. — Diagrammatic representation of the plankton in a Wisconsin lake during 
May to October, the numbers being those of organisms per litre. R = Rotifers ; 
N = Nauplii ; C = Crustacea ; M = depth in metres ; A = Algae other than 
Diatoms ; D = Diatoms ; O, == dissolved oxygen in cubic centimetres per litre; 
T = temperature in degrees Centigrade ; tr. = trace. (After Welch.) 

where the latter acts as a brake on eddy-diffusion currents and hence 
allows depletion by sinking. There may also be marked differences 
on windy and still days, as indicated in Fig. 166 of the representation 
of a Blue-green Alga of low specific gravity, which accumulates at 



the surface in calm weather and is then absent below 4 metres (a), 
but in windy weather is plentiful down to much greater depths (b). 
Yet other Blue-green Algae, such as Aphanizomenon flos-aquae, can 
achieve a vertical distribution of 6 metres or more on very calm days 
(according to Professor G. W. Prescott in litt.). However, the really 
lasting differences in composition occur in and below the thermo- 
cline, where the eddy-diffusion currents are curtailed. The latter 
may, however, extend to considerable depths during the autumn 

Fig. 166. — Diagram indicating distribution in a European lake of a cyanophycean 
{Glosotrichia echinulata), which is rendered buoyant by included gas- vacuoles : 
(a) during calm weather, and (b) during windy weather. (After Ruttner, modified.) 

circulation, and at such times lead to a virtually uniform distribution 
of plankton (cf. the October section of Fig. 165). Later on, under 
the winter ice-cover of cool to frigid regions, stratification again 
appears. Another complicating physical factor influencing stratifica- 
tion is wave action, and yet another is water renewal, during which 
much of the plankton-rich surface water of lakes is liable to be lost 
by outflow and commonly replaced by inflowing river or other water 
poor in plankton. 

The biotic factors affecting phytoplankton are far more numerous 
and complicated than the mechanical ones, and no attempt will be 
made to analyse them here. They are concerned with such (often 
interrelated) processes as reproduction, photosynthesis, secretion of 


external 'envelopes,' parasitism, 'grazing 1 ' by animals, and active 
movement, as well as their various components. All of these can 
be governed in turn by temperature, light, or the chemical composition 
of the water as already mentioned. It should be emphasized, how- 
ever, that temperature differences are less important than many 
others, being often superseded by such factors as light, the effect of 
which on photosynthesis in the twilight region (dysphoric zone) is 
nevertheless in turn influenced by temperature. The long wave- 
length ' red ' radiation is absorbed in the upper layers so that at a 
depth of commonly 15-20 metres in clear water a vivid ' green ' 
coloration predominates where there are objects to reflect the rays. 
Absorption and subsequent use for photosynthesis being largelv 
complementary to the colours of plants, the light of these and of 
greater depths is utilized best by phytoplanktonic organisms that 
are brownish (such as Diatoms) or reddish (such as the flagellate 
Rhodomonas and certain Cyanophyceae). These brownish and reddish 
types are often predominant in deep fresh waters, as are Brown and 
particularly Red Algae in the sea. 

Owing apparently to wind and wave action as well as to the 
mysterious avoidance of shallow water by Entomostraca, the character 
of the general plankton is often peculiar along lake-shores, where the 
abundance of phytoplankton may actually be greater than elsewhere 
owing to turbulence and the low incidence of predation. On the 
other hand, in deep waters where oxygen tends to be in short supplv, 
the temperature is often so low that few organisms can exist, and, 
therefore, much less of this gas than usual is required for respiratio